ARTICLE | doi:10.20944/preprints201811.0546.v4
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Convolutional Neural Network (CNN), Deep learning, Architecture, Applications
Online: 14 February 2019 (10:01:31 CET)
With the increase of the Artificial Neural Network (ANN), machine learning has taken a forceful twist in recent times. One of the most spectacular kinds of ANN design is the Convolutional Neural Network (CNN). The Convolutional Neural Network (CNN) is a technology that mixes artificial neural networks and up to date deep learning strategies. In deep learning, Convolutional Neural Network is at the center of spectacular advances. This artificial neural network has been applied to several image recognition tasks for decades and attracted the eye of the researchers of the many countries in recent years as the CNN has shown promising performances in several computer vision and machine learning tasks. This paper describes the underlying architecture and various applications of Convolutional Neural Network.
ARTICLE | doi:10.20944/preprints202206.0414.v2
Subject: Computer Science And Mathematics, Computational Mathematics Keywords: Elasticity; Machine learning; Minimum potential energy; Partial differential equations (PDEs); Physics-informed neural network
Online: 3 April 2023 (03:29:13 CEST)
The deep energy method (DEM), a type of physics-informed neural network, is evolving as an alternative to finite element analysis. This method employs the principle of minimum potential energy to predict deformations under static loading conditions. However, the model’s accuracy is contingent upon choosing the appropriate architecture for the model, which can be challenging due to the high interactions between hyperparameters, large search space, difficulty in identifying objective functions, and non-convex relationships with the objective functions. To improve DEM’s accuracy, we first introduce random Fourier feature (RFF) mapping. RFF mapping helps with the training of the model by reducing bias towards high frequencies. The effects of six hyperparameters are then studied under compression, tension, and bending loads in planar linear elasticity. Based on this study, a systematic automated hyperparameter optimization approach is proposed. Due to the high interaction between hyperparameters and the non-convex nature of the optimization problem, Bayesian optimization algorithms are used. The models trained using optimized hyperparameters and having Fourier feature mapping can accurately predict deflections compared to finite element analysis. Additionally, the deflections obtained for tension and compression load cases are more sensitive to variations in hyperparameters than bending.
ARTICLE | doi:10.20944/preprints202107.0252.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: Recurrent neural network; Long-term short memory; Gated recurrent unit
Online: 12 July 2021 (12:03:06 CEST)
Deep neural networks (DNNs) have made a huge impact in the field of machine learning by providing unbeatable humanlike performance to solve real-world problems such as image processing and natural language processing (NLP). Convolutional neural network (CNN) and recurrent neural network (RNN) are two typical architectures that are widely used to solve such problems. Time sequence-dependent problems are generally very challenging, and RNN architectures have made an enormous improvement in a wide range of machine learning problems with sequential input involved. In this paper, different types of RNN architectures are compared. Special focus is put on two well-known gated-RNN’s Long Term Short Memory (LSTM) and Gated Recurrent Unit (GRU). We evaluated these models on the task of force estimation system in pouring. In this study, four different models including multi-layers LSTM, multi-layers GRU, single-layer LSTM and single-layer GRU) were created and trained. The result suggests that multi-layer GRU outperformed other three models.
ARTICLE | doi:10.20944/preprints202002.0318.v2
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: activity recognition; time series classification; neural; networks; deep learning; machine learning; CNNs; LSTMs; many-to-many
Online: 15 April 2020 (08:02:43 CEST)
We present and benchmark FilterNet, a flexible deep learning architecture for time series classification tasks, such as activity recognition via multichannel sensor data. It adapts popular CNN and CNN-LSTM motifs which have excelled in activity recognition benchmarks, implementing them in a many-to-many architecture to markedly improve frame-by-frame accuracy, event segmentation accuracy, model size, and computational efficiency. We propose several model variants, evaluate them alongside other published models using the Opportunity benchmark dataset, demonstrate the effect of model ensembling and of altering key parameters, and quantify the quality of the models’ segmentation of discrete events. We also offer recommendations for use and suggest potential model extensions. FilterNet advances the state of the art in all measured accuracy and speed metrics on the benchmarked dataset, and it can be extensively customized for other applications.
ARTICLE | doi:10.20944/preprints202101.0133.v1
Subject: Engineering, Automotive Engineering Keywords: Energy-efficiency; HVAC system; Neural network; Cooling load; Metaheuristic search.
Online: 8 January 2021 (10:20:07 CET)
Early prediction of thermal loads plays an essential role in analyzing energy-efficient buildings' energy performance. On the other hand, stochastic algorithms have recently shown high proficiency in dealing with this issue. These are the reasons that this work is dedicated to evaluating an innovative hybrid method for predicting the cooling load (CL) in buildings with residential usage. The proposed model is a combination of artificial neural networks and stochastic fractal search (SFS-ANN). Two benchmark algorithms, namely the grasshopper optimization algorithm (GOA) and firefly algorithm (FA), are also considered to be compared with the SFS. The non-linear effect of eight independent factors on the CL is analyzed using each model's optimal structure. Evaluation of the results outlined that all three metaheuristic algorithms (with more than 90 % correlation) can adequately optimize the ANN. In this regard, this tool's prediction error declined by nearly 23, 18, and 36 % by applying the GOA, FA, and SFS techniques. Also, all used accuracy criteria indicated the superiority of the SFS over the benchmark schemes. Therefore, it is inferred that utilizing the SFS along with ANN provides a reliable hybrid model for the early prediction of CL.
ARTICLE | doi:10.20944/preprints202101.0411.v1
Subject: Engineering, Automotive Engineering Keywords: Energy performance; Cooling load prediction; Neural network, Metaheuristic optimization.
Online: 21 January 2021 (09:23:04 CET)
Regarding the high efficiency of metaheuristic techniques in energy performance analysis, this paper scrutinizes and compares five novel optimizers, namely biogeography-based optimization (BBO), invasive weed optimization (IWO), social spider algorithm (SOSA), shuffled frog leaping algorithm (SFLA), and harmony search algorithm (HSA) for the early prediction of cooling load in residential buildings. The algorithms are coupled with a multi-layer perceptron (MLP) to adjust the neural parameters that connect the CL with the influential factors. The complexity of the models is optimized by means of a trial-and-error effort, and it was shown that the BBO and IWO need more crowded spaces for fulfilling the optimization. The results revealed that the internal parameters (i.e., biases and weights) suggested by the BBO generate the most reliable MLP for both analyzing and generalizing the CL pattern (with nearly 93 and 92% correlations, respectively). Followed by this, the IWO emerged as the second powerful optimizer with mean absolute errors of 1.8632 and 1.9110 in the training and testing phases. Therefore, the BBO-MLP and IWO-MLP can be reliably used for accurate analysis of the CL in future projects.
ARTICLE | doi:10.20944/preprints202104.0501.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: convolutional neural networks; dilated neural networks; optimality
Online: 19 April 2021 (15:00:30 CEST)
One of the most effective image processing techniques is the use of convolutional neural networks, where we combine intensity values at grid points in the vicinity of each point. To speed up computations, researchers have developed a dilated version of this technique, in which only some points are processed. It turns out that the most efficient case is when we select points from a sub-grid. In this paper, we explain this empirical efficiency proving that the sub-grid is indeed optimal – in some reasonable sense. To be more precise, we prove that all reasonable optimality criteria, the optimal subset of the original grid is either a sub-grid, or a sub-grid-like set.
ARTICLE | doi:10.20944/preprints202011.0233.v1
Subject: Physical Sciences, Optics And Photonics Keywords: microcombs; neural networks; optical neural networks; photonics
Online: 6 November 2020 (09:19:13 CET)
Optical artificial neural networks (ONNs) have significant potential for ultra-high computing speed and energy efficiency. We report a new approach to ONNs based on integrated Kerr micro-combs that is programmable, highly scalable and capable of reaching ultra-high speeds, demonstrating the building block of the ONN — a single neuron perceptron — by mapping synapses onto 49 wavelengths to achieve a single-unit throughput of 11.9 Giga-OPS at 8 bits per OP, or 95.2 Gbps. We test the perceptron on handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. By scaling the perceptron to a deep learning network using off-the-shelf telecom technology we can achieve high throughput operation for matrix multiplication for real-time massive data processing.
ARTICLE | doi:10.20944/preprints202306.0454.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: intracranial pressure; cerebral compliance; deep neural networks; recurrent neural networks; convolutional neural networks
Online: 6 June 2023 (11:50:05 CEST)
The intracranial pressure (ICP) signal, as monitored on patients in intensive care units, contains pulses of cardiac origin where P1 and P2 subpeaks can often be observed. When calculable, the ratio of their relative amplitudes is an indicator of the patient’s cerebral compliance. This characterization is particularly informative on the overall state of the cerebrospinal system. We developed a recurrent neural network-based framework for P2/P1 ratio computation that only takes a raw ICP signal as an input. Two tasks are performed, namely pulse classification and subpeak designation. Pulse classification was achieved with an area under the curve of 0.90 on a 4,344-pulse testing dataset, while the peak designation algorithm identified pulses with a P2/P1 ratio > 1 with a 97.3% accuracy. Although it still needs to be evaluated on a larger number of labeled recordings, our automated P2/P1 ratio calculation framework appears to be a promising tool for improving bedside monitoring devices.
ARTICLE | doi:10.20944/preprints202306.0671.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Earthquake; Recurrent neural network; Prediction; Artificial neural network
Online: 9 June 2023 (05:21:52 CEST)
An earthquake is a natural event by its general definition. This natural event is a disaster that causes significant damage, loss of life, and other economic effects that will damage the state. The possibility of predicting a natural event such as an earthquake will minimize the reasons mentioned. Data collection, data processing, and data evaluation were carried out in this study. Earthquake forecasting was performed using the data and the RNN (Recurrent Neural Network) method. The study was carried out on seismic data with a magnitude of 3.0 and above belonging to Düzce Province between 1990 and 2022. In order to increase the learning potential of the method, the b and d values of the earthquake were calculated and included in the data set, except for the earthquake magnitude. The determination of earthquakes in a specific time interval in regions of Turkey, the classification of earthquake-related seismic data using artificial neural networks, and the production of predictions for the future reveal the importance of this study.
ARTICLE | doi:10.20944/preprints202305.1609.v1
Subject: Engineering, Control And Systems Engineering Keywords: neural dynamics; neural oscillation; bio-inspiration; artificial intelligence
Online: 23 May 2023 (08:05:36 CEST)
This paper investigates the dynamic properties of artificial neural networks using differential equations and explores the influence of parameters on stability and neural oscillations. By analyzing the equilibrium point of the differential equations, we identify conditions for asymptotic stability and criteria for oscillation in artificial neural networks. Furthermore, we demonstrate how adjusting synaptic weights between neurons can effectively control stability and oscillation. The proposed model offers potential insights into the malfunctioning mechanisms of biological neural networks implicated in neurological disorders like Parkinson's disease tremors and epilepsy seizures, which are characterized by abnormal oscillations.
ARTICLE | doi:10.20944/preprints201901.0319.v1
Subject: Chemistry And Materials Science, Nanotechnology Keywords: cascaded neural networks; memristor crossbar; convolutional neural networks
Online: 31 January 2019 (06:54:33 CET)
Multiply-accumulate calculations using a memristor crossbar array is an important method to realize neuromorphic computing. However, the memristor array fabrication technology is still immature, and it is difficult to fabricate large-scale arrays with high-yield, which restricts the development of memristor-based neuromorphic computing technology. Therefore, cascading small-scale arrays to achieve the neuromorphic computational ability that can be achieved by large-scale arrays, which is of great significance for promoting the application of memristor-based neuromorphic computing. To address this issue, we present a memristor-based cascaded framework with some basic computation units, several neural network processing units can be cascaded by this means to improve the processing capability of the dataset. Besides, we introduce a split method to reduce pressure of input terminal. Compared with VGGNet and GoogLeNet, the proposed cascaded framework can achieve 93.54% Fashion-MNIST accuracy under the 4.15M parameters. Extensive experiments with Ti/AlOx/TaOx/Pt we fabricated are conducted to show that the circuit simulation results can still provide a high recognition accuracy, and the recognition accuracy loss after circuit simulation can be controlled at around 0.26%.
REVIEW | doi:10.20944/preprints202012.0122.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: neural stemness; neural stem/progenitor cell; tumor-initiating cell; neural ground state; neural default model; differentiation potential; tumorigenicity; tumorigenesis; evo-devo
Online: 7 December 2020 (07:02:34 CET)
Tumorigenesis is a complex biological phenomenon that includes extensive genetic and phenotypic heterogeneities and complicated regulatory mechanisms. In the recent few years, our studies demonstrate that tumor-initiating cells are similar to neural stem/progenitor cells in regulatory networks, tumorigenicity and pluripotent differentiation potential. In the review, I will make further discussion on these observations and propose a rule of cell biology by integrating these findings with evidence from developmental biology, tumor biology and evolution, which suggests that neural stemness underlies two coupled cell properties, tumorigenicity and pluripotent differentiation potential. Tumorigenicity and phenotypic heterogeneity in tumor is a result of acquirement of neural stemness in cells. The neural stemness property of tumor-initiating cells can hopefully integrate different concepts/hypotheses underlying tumorigenesis. Neural stem cells/neural progenitors and tumor-initiating cells share regulatory networks; both exhibit neural stemness, tumorigenicity and differentiation potential; both are dependent on expression or activation of ancestral genes (the atavistic effect); both rely primarily on aerobic glycolytic metabolism; both can differentiate into various cells or tissues that are derived from three germ layers, resembling severely disorganized or more severely degenerated process of embryonic development; both are enriched in long genes with more splice variants that provide more plastic scaffolds for cell differentiation, etc. The property of neural stemness might be a key point to understand tumorigenesis and pluripotent differentiation potential, and possibly explain certain pathological observations in tumors that have been inexplicable. Therefore, behind the complexity of tumorigenesis might be a general rule of cell biology, i.e., neural stemness represents the ground state of cell tumorigenicity and pluripotent differentiation potential.
ARTICLE | doi:10.20944/preprints202102.0549.v1
Subject: Engineering, Automotive Engineering Keywords: microcombs; neural networks
Online: 24 February 2021 (12:55:28 CET)
Convolutional neural networks (CNNs), inspired by biological visual cortex systems, are a powerful category of artificial neural networks that can extract the hierarchical features of raw data to greatly reduce the network parametric complexity and enhance the predicting accuracy. They are of significant interest for machine learning tasks such as computer vision, speech recognition, playing board games and medical diagnosis [1-7]. Optical neural networks offer the promise of dramatically accelerating computing speed to overcome the inherent bandwidth bottleneck of electronics. Here, we demonstrate a universal optical vector convolutional accelerator operating beyond 10 Tera-OPS (TOPS - operations per second), generating convolutions of images of 250,000 pixels with 8-bit resolution for 10 kernels simultaneously — enough for facial image recognition. We then use the same hardware to sequentially form a deep optical CNN with ten output neurons, achieving successful recognition of full 10 digits with 900 pixel handwritten digit images with 88% accuracy. Our results are based on simultaneously interleaving temporal, wavelength and spatial dimensions enabled by an integrated microcomb source. This approach is scalable and trainable to much more complex networks for demanding applications such as unmanned vehicle and real-time video recognition.
Subject: Biology And Life Sciences, Cell And Developmental Biology Keywords: Neural Stem Cell; Secretome; Neurodevelopment; Radial Glia; Neural Progenitor Cell
Online: 4 October 2021 (12:25:30 CEST)
Neural stem cell (NSC) based therapies are at the forefront of regenerative medicine strategies to combat illness and injury of the central nervous system (CNS). In addition to their ability to produce new cells, NSCs secrete a variety of products, known as the NSC secretome, that have been shown to ameliorate CNS disease pathology and promote recovery. As pre-clinical and clinical research to harness the NSC secretome for therapeutic purposes advances, a more thorough understanding of the endogenous NSC secretome can provide useful insight into the functional capabilities of NSCs. In this review, we focus on research investigating the autocrine and paracrine functions of the endogenous NSC secretome across life. We also compare the NSC secretome across species, finding signs of conserved parallels between rodent, human and zebrafish NSC secretomes. Throughout development and adulthood, we find evidence that the NSC secretome is a critical component of how endogenous NSCs regulate themselves and their niche. We also find gaps in current literature, most notably in the clinically relevant domain of endogenous NSC paracrine function in the injured CNS. Future investigations to further define the endogenous NSC secretome and its role in CNS tissue regulation are necessary to bolster our understanding of NSC-niche interactions and to aid in the generation of safe and effective NSC-based therapies.
BRIEF REPORT | doi:10.20944/preprints202210.0208.v1
Subject: Computer Science And Mathematics, Information Systems Keywords: Quantum neural network; Breast cancer; Classical neural network; Machine learning; Mammography
Online: 14 October 2022 (10:11:54 CEST)
Computer-aided image diagnostics (CAD) have been used in many fields of diagnostic medicine. It relies heavily on classical computer vision and artificial intelligence. Quantum neural network (QNN) has been introduced by many researchers around the world and presented recently by research corporations such as Microsoft, Google, and IBM. In this paper, the investigation of the validity of using the QNN algorithm for machine-based breast cancer detection was performed. To validate the learnability of the QNN, a series of learnability tests were performed alongside with classical convolutional neural network (CCNN). QNN is built using the Cirq library to perform the assimilation of quantum computation on classical computers. Series of investigations were performed to study the learnability characteristics of QNN and CCNN under the same computational conditions. The comparison was performed for real Mammogram data sets. The investigations showed success in terms of recognizing the data and training. Our work shows better performance of QNN in terms of successfully training and producing a valid model for smaller data set compared to CCNN.
Subject: Engineering, Electrical And Electronic Engineering Keywords: optical neural network; microcomb
Online: 12 March 2020 (03:41:39 CET)
Optical artificial neural networks (ONNs) — analog computing hardware tailored for machine learning [1,2] — have significant potential for ultra-high computing speed and energy efficiency . We propose a new approach to architectures for ONNs based on integrated Kerr micro-comb sources  that is programmable, highly scalable and capable of reaching ultra-high speeds. We experimentally demonstrate the building block of the ONN — a single neuron perceptron — by mapping synapses onto 49 wavelengths of a micro-comb to achieve a high single-unit throughput of 11.9 Giga-FLOPS at 8 bits per FLOP, corresponding to 95.2 Gbps. We test the perceptron on simple standard benchmark datasets — handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. This performance is a direct result of the record small wavelength spacing (49GHz) for a coherent integrated microcomb source, which results in an unprecedented number of wavelengths for neuromorphic optics. Finally, we propose an approach to scaling the perceptron to a deep learning network using the same single micro-comb device and standard off-the-shelf telecommunications technology, for high-throughput operation involving full matrix multiplication for applications such as real-time massive data processing for unmanned vehicle and aircraft tracking.
ARTICLE | doi:10.20944/preprints202305.1490.v1
Subject: Engineering, Civil Engineering Keywords: Surrogate Model; Convolutional Neural Network; Physics-Informed Neural Networks; Elliptic PDE; FEM
Online: 22 May 2023 (09:48:22 CEST)
This study aimed at exploring what role artificial intelligence techniques could play in the futural numerical analysis. In this paper, a convolutional neural network techniques based on modified loss function is proposed as a surrogate of finite element method(FEM). Several surrogate-based physics-informed neural networks(PINNs) are developed to solve representative boundary value problems based on elliptic partial differential equations (PDEs). Results from the proposed surrogate-based approach are in good agreement with ones from conventional FEM. It is found that modification of the loss function could improve the prediction accuracy of the neural network. It is indicated that to some extent the artificial intelligence technique could replace conventional numerical analysis as a great surrogate model.
Subject: Engineering, Automotive Engineering Keywords: forest fire; image recognition; graph neural network;
Online: 13 July 2021 (11:31:18 CEST)
Forest fire identification is important for forest resource protection. Effective monitoring of forest fires requires the deployment of multiple monitors with different viewpoints, while most traditional recognition models can only recognize images from a single source. By ignoring the information from images with different viewpoints, these models produce high rates of missed and false alarms. In this paper, we propose a graph neural network model based on the similarity of dynamic features of multi-view images to improve the accuracy of forest fire recognition. The input features of the nodes on the graph are converted into relational features of different gallery pairs by establishing pairs (nodes) representing different viewpoint images and gallery images. The new feature library relationship is used to update the image gallery with dynamic features in order to achieve the estimation of similarity between images and improve the image recognition rate of the model. In addition, to reduce the complexity of image pre-processing process and extract key features in images effectively, this paper also proposes a dynamic feature extraction method for fire regions based on image segment ability. By setting the threshold value of HSV color space, the fire region is segmented from the image, and the dynamic features of successive frames of the fire region are extracted. The experimental results show that, compared with the baseline method Resnet, this paper's method is more effective in identifying forest fires, and its recognition accuracy is improved by 2%. And the scheme of this paper can adapt to different forest fire scenes, with better generalization ability and anti-interference ability.
REVIEW | doi:10.20944/preprints202208.0203.v2
Subject: Biology And Life Sciences, Cell And Developmental Biology Keywords: neural induction; embryogenesis; tumorigenesis; neural stemness; tumorigenicity; pluripotency; epithelial-mesenchymal transition; tumor microenvironment
Online: 9 March 2023 (06:57:01 CET)
Characterization of cancer cells and neural stem cells indicates that tumorigenicity and pluripotency are coupled cell properties determined by neural stemness, and tumorigenesis represents a process of progressive loss of original cell identity and gain of neural stemness. This reminds of a most fundamental process required for the development of the nervous system and body axis during embryogenesis, i.e., embryonic neural induction. Neural induction is that, in response to extracellular signals that are secreted by the Spemann-Mangold organizer in amphibians or the node in mammals and inhibit epidermal fate in ectoderm, the ectodermal cells lose their epidermal fate and assume the neural default fate and consequently, turn into neuroectodermal cells. They further differentiate into the nervous system and also some non-neural cells via interaction with adjacent tissues. Failure in neural induction leads to failure of embryogenesis, and ectopic neural induction due to ectopic organizer or node activity or activation of embryonic neural genes causes a formation of secondary body axis or a conjoined twin. During tumorigenesis, cells progressively lose their original cell identity and gain of neural stemness, and consequently, gain of tumorigenicity and pluripotency, due to various intra-/extracellular insults in cells of a postnatal animal. Tumorigenic cells can be induced to differentiation into normal cells and integrate into normal embryonic development within an embryo. However, they form tumors and cannot integrate into animal tissues/organs in a postnatal animal because of lack of embryonic inducing signals. Combination of studies of developmental and cancer biology indicates that neural induction drives embryogenesis in gastrulating embryos but a similar process drives tumorigenesis in a postnatal animal. Tumorigenicity is the manifestation of aberrant occurrence of pluripotent state in a postnatal animal. Pluripotency and tumorigenicity are both but different manifestations of neural stemness in pre- and postnatal stage, respectively, of animal life. The unique property of neural stemness is derived from the evolutionary advantage of neural genes and the neural-biased state of the last common unicellular ancestors of metazoan. Based on these findings, I discuss about some confusion in cancer research, propose to distinguish the causality and associations and discriminate causal and supporting factors involved in tumorigenesis, and suggest revisiting the focus of cancer research.
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: microcombs; optical neural networks; neuromorphic computing, artificial intelligence; Kerr microcombs; convolutional neural network
Online: 16 November 2020 (13:30:14 CET)
Convolutional neural networks (CNNs), inspired by biological visual cortex systems, are a powerful category of artificial neural networks that can extract the hierarchical features of raw data to greatly reduce the network parametric complexity and enhance the predicting accuracy. They are of significant interest for machine learning tasks such as computer vision, speech recognition, playing board games and medical diagnosis [1-7]. Optical neural networks offer the promise of dramatically accelerating computing speed to overcome the inherent bandwidth bottleneck of electronics. Here, we demonstrate a universal optical vector convolutional accelerator operating beyond 10 Tera-FLOPS (floating point operations per second), generating convolutions of images of 250,000 pixels with 8-bit resolution for 10 kernels simultaneously — enough for facial image recognition. We then use the same hardware to sequentially form a deep optical CNN with ten output neurons, achieving successful recognition of full 10 digits with 900 pixel handwritten digit images with 88% accuracy. Our results are based on simultaneously interleaving temporal, wavelength and spatial dimensions enabled by an integrated microcomb source. This approach is scalable and trainable to much more complex networks for demanding applications such as unmanned vehicle and real-time video recognition.
REVIEW | doi:10.20944/preprints202304.0648.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: optimization methods; physics-informed neural networks; spiking neural networks; quantum neural networks; graph neural networks; information geometry; quasi-Newton methods; approximation; quantum computations; gradient free optimization; fractional order optimization; bilevel optimization
Online: 20 April 2023 (10:12:27 CEST)
Creating self-learning algorithms, developing deep neural networks and improving other methods that "learn" for various areas of human activity is the main goal of the theory of machine learning. It helps to replace the human with a machine, aiming to increase the quality of production. The theory of artificial neural networks, which already have replaced the humans in problems of detection of moving objects, recognition of images or sounds, time series prediction, big data analysis and numerical methods remains the most dispersed branch of the theory of machine learning. Certainly, for each area of human activity it is necessary to select appropriate neural network architectures, methods of data processing and some novel tools from applied mathematics. But the universal problem for all these neural networks with specific data is the achieving the highest accuracy in short time. Such problem can be resolved by increasing sizes of architectures and improving data preprocessing, where the accuracy rises with the training time. But there is a possibility to increase the accuracy without time growing, applying optimization methods. In this survey we demonstrate existing optimization algorithms of all types, which can be used in neural networks. There are presented modifications of basic optimization algorithms, such as stochastic gradient descent, adaptive moment estimation, Newton and quasi-Newton optimization methods. But the most recent optimization algorithms are related to information geometry, for Fisher-Rao and Bregman metrics. This approach in optimization extended the theory of classic neural networks to quantum and complex-valued neural networks, due to geometric and probabilistic tools. There are provided applications of all introduced optimization algorithms, what delighted many kinds of neural networks, which can be improved by including any advanced approaches in minimization of the loss function. Afterwards, we demonstrated ways of developing optimization algorithms in further researches, engaging neural networks with progressive architectures. Classical gradient based optimizers can be replaced by fractional order, bilevel and, even, gradient free optimization methods. There is a possibility to add such analogues in graph, spiking, complex-valued, quantum and wavelet neural networks. Besides the usual problems of image recognition, time series prediction, object detection, there are many are other tasks for modern theory of machine learning, such as solving problem of quantum computations, partial differential and integro-differential equations, stochastic processes and Brownian motion, making decisions and computer algebra.
ARTICLE | doi:10.20944/preprints202303.0001.v3
Subject: Computer Science And Mathematics, Computational Mathematics Keywords: Artificial neural networks; back-propagation
Online: 9 March 2023 (02:04:21 CET)
We present a simplified computational rule for the back-propagation formulas for artificial neural networks. In this work, we provide a generic two-step rule for the back-propagation algorithm in matrix notation. Moreover, this rule incorporates both the forward and backward phases of the computations involved in the learning process. Specifically, this recursively computing rule permits the propagation of the changes to all synaptic weights in the network, layer by layer, efficiently. In particular, we use this rule to compute both the up and down partial derivatives of the cost function of all the connections feeding into the output layer.
ARTICLE | doi:10.20944/preprints202209.0231.v1
Subject: Computer Science And Mathematics, Probability And Statistics Keywords: neural networks; regularization; deep networks
Online: 15 September 2022 (13:06:13 CEST)
Numerous approaches address over-fitting in neural networks: by imposing a penalty on the parameters of the network (L1, L2, etc); by changing the network stochastically (drop-out, Gaussian noise, etc.); or by transforming the input data (batch normalization, etc.). In contrast, we aim to ensure that a minimum amount of supporting evidence is present when fitting the model parameters to the training data. This, at the single neuron level, is equivalent to ensuring that both sides of the separating hyperplane (for a standard artificial neuron) have a minimum number of data points — noting that these points need not belong to the same class for the inner layers. We firstly benchmark the results of this approach on the standard Fashion-MINST dataset, comparing it to various regularization techniques. Interestingly, we note that by nudging each neuron to divide, at least in part, its input data, the resulting networks make use of each neuron, avoiding a hyperplane completely on one side of its input data (which is equivalent to a constant into the next layers). To illustrate this point, we study the prevalence of saturated nodes throughout training, showing that neurons are activated more frequently and earlier in training when using this regularization approach. A direct consequence of the improved neuron activation is that deep networks are now easier to train. This is crucially important when the network topology is not known a priori and fitting often remains stuck in a suboptimal local minima. We demonstrate this property by training a network of increasing depth (and constant width): most regularization approaches will result in increasingly frequent training failures (over different random seeds) whilst the proposed evidence-based regularization significantly outperforms in its ability to train deep networks.
ARTICLE | doi:10.20944/preprints202103.0676.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: Neural network; MET; exon skipping
Online: 26 March 2021 (16:29:10 CET)
Background: Disruption of alternative splicing (AS) is frequently observed in cancer and it might represent an important signature for tumor progression and therapy. Exon skipping (ES) represents one of the most frequent AS events and in non-small cell lung cancer (NSCLC) MET exon 14 skipping was shown to be targetable. Methods: We constructed a neural network (NN) specifically designed to detect MET exon 14 skipping events using RNAseq data. Furthermore, for discovery purpose we also developed a sparsely connected autoencoder to identify uncharacterized MET isoforms. Results: The NN had 100% Met exon 14 skipping detection rate, when tested on a manually curated set of 690 TCGA bronchus and lung samples. When globally applied to 2605 TCGA samples, we observed that the majority of false positives was characterized by a blurry coverage of exon 14, but interesting they share a common coverage peak in the second intron and we speculate that this event could be the transcription signature of a LINE1-MET fusion. Conclusions: Taken together our results indicate that neural networks can be an effective tool to provide a quick classification of pathological transcription events and sparsely connected autoencoders could represent the basis for the development of an effective discovery tool.
ARTICLE | doi:10.20944/preprints202103.0033.v1
Subject: Engineering, Automotive Engineering Keywords: microcombs; optical neural networks; perceptron
Online: 1 March 2021 (14:56:09 CET)
Optical artificial neural networks (ONNs) have significant potential for ultra-high computing speed and energy efficiency. We report a novel approach to ONNs that uses integrated Kerr optical micro-combs. This approach is programmable and scalable and is capable of reaching ultra-high speeds. We demonstrate the basic building block ONNs — a single neuron perceptron — by mapping synapses onto 49 wavelengths to achieve an operating speed of 11.9 x 109 operations per second, or Giga-OPS, at 8 bits per operation, which equates to 95.2 gigabits/s (Gbps). We test the perceptron on handwritten-digit recognition and cancer-cell detection — achieving over 90% and 85% accuracy, respectively. By scaling the perceptron to a deep learning network using off-the-shelf telecom technology we can achieve high throughput operation for matrix multiplication for real-time massive data processing.
ARTICLE | doi:10.20944/preprints202005.0521.v1
Subject: Computer Science And Mathematics, Probability And Statistics Keywords: Multiplicative error; ARIMA; Neural net
Online: 31 May 2020 (21:50:01 CEST)
Real-world time series data sets contain a combination of linear and nonlinear patterns, making the time series forecasting problem more challenging. In this paper, a new hybrid methodology is introduced for forecasting univariate time series data sets using a multiplicative error modeling approach. An autoregressive integrated moving average (ARIMA) model is combined with an autoregressive neural network (ARNN) for improving the predictions of individual forecast models. The proposed multiplicative ARIMA-ARNN model glorifies the chances of capturing the different combinations of linear and nonlinear patterns in time series. The model shows outstanding performance on six standard time-series data sets compared to other widely used single and hybrid forecasting models.
ARTICLE | doi:10.20944/preprints201808.0034.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: android; malware; convolutional neural network
Online: 2 August 2018 (06:12:48 CEST)
Using smartphone especially android platform has already got eighty percent market shares, due to aforementioned report, it becomes attacker’s primary goal. There is a growing number of private data onto smart phones and low safety defense measure, attackers can use multiple way to launch and to attack user’s smartphones.(e.g. Using different coding style to confuse the software of detecting malware). Existing android malware detection methods use multiple features, like safety sensor API, system call, control flow structure and data information flow, then using machine learning to check whether its malware or not. These feature provide app’s unique property and limitation, that is to say, from some perspectives it might suit for some specific attack, but wouldn’t suit for others. Nowadays most malware detection methods use only one aforementioned feature, and these methods mostly analysis to detect code, but facing the influence of malware’s code confusion and zero-day attack, aforementioned feature extraction method may cause wrong judge. So, it’s necessary to design an effective technique analysis to prevent malware. In this paper, we use the importance of word from apk, because of code confusion, some malware attackers only rename variables, if using general static analysis wouldn’t judge correctly, then use these importance value to go through our proposed method to generate picture, finally using convolutional neural network to see whether the apk file is malware or not.
ARTICLE | doi:10.20944/preprints202305.0178.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Shallow neural networks; recurrent neural networks; predictive hybrid model; photovoltaic energy; photovoltaic energy prediction
Online: 4 May 2023 (03:53:12 CEST)
This article presents a forecast model that uses a hybrid architecture of recurrent neural networks (RNN) with surface neural networks (ANN), based on historical records of exported active energy (EAE) and weather data. Two types of models were developed: the first type includes six models that use EAE records and weather variables as inputs, while the second type includes eight models that use only weather variables. Different metrics were applied to assess the performance of these models, and the best model of each type was selected. Finally, a comparison of the performance between the selected models of both types is presented, and they are validated with real data provided by a solar plant, achieving acceptable levels of accuracy. The selected model of the first type has an RMSE of 0.19, MSE of 0.03, MAE of 0.09, a correlation coefficient of 0.96, and a determination coefficient of 0.93. The other selected model of the second type showed lower precision in the metrics (RMSE = 0.24, MSE = 0.06, MAE = 0.10, Corr. Coef. = 0.95, and Det. Coef. = 0.90). Both models demonstrated good performance and acceptable accuracy in forecasting the weekly photovoltaic energy production of the solar plant.
ARTICLE | doi:10.20944/preprints202207.0021.v1
Subject: Biology And Life Sciences, Biology And Biotechnology Keywords: artificial neural networks; biological neural networks; cortical prosthetic vision; machine vision; neuromorphic hardware; neuroprosthesis
Online: 1 July 2022 (17:01:32 CEST)
Sense element engagement theory explains how neural networks produce cortical prosthetic vision. A major prediction of the theory can be tested by developing a device which is expected to enable perception of continuous forms in altered visual geometries. The research reported here completes several essential steps in developing this device: (1) replication of simulations that are consistent with the theory using the NEST simulator, which can also be used for full-scale network emulation by a neuromorphic computer; (2) testing whether results consistent with the theory survive increasing the scale and duration of simulations; (3) establishing a method that uses numbers of spikes produced by network neurons to report the number of phosphenes produced by cortical stimulation; and (4) simulating essential functions of the prosthetic device. NEST simulations replicated early results and increasing their scale and duration produced results consistent with the theory. A decision function created using multinomial logistic regression correctly classified the expected number of phosphenes for 2080 spike number distributions for each of three sets of data, half of which arise from simulations expected to yield continuous visual forms on an altered visual geometry. A process for modulating electrical stimulation amplitude based on intermittent population recordings that is predicted to produce continuous visual forms was successfully simulated. The classification function developed using logistic regression will be used to tune this process as the scale of simulations is further increased.
ARTICLE | doi:10.20944/preprints201907.0121.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Artificial Neural Networks; Deep Learning; Generative Neural Networks; Incremental Learning; Novelty detection; Catastrophic Interference
Online: 8 July 2019 (14:29:28 CEST)
Deep learning models are part of the family of artificial neural networks and, as such, it suffers of catastrophic interference when they learn sequentially. In addition, most of these models have a rigid architecture which prevents the incremental learning of new classes. To overcome these drawbacks, in this article we propose the Self-Improving Generative Artificial Neural Network (SIGANN), a type of end-to-end Deep Neural Network system which is able to ease the catastrophic forgetting problem when leaning new classes. In this method, we introduce a novelty detection model to automatically detect samples of new classes, moreover an adversarial auto-encoder is used to produce samples of previous classes. This system consists of three main modules: a classifier module implemented using a Deep Convolutional Neural Network, a generator module based on an adversarial autoencoder; and a novelty detection module, implemented using an OpenMax activation function. Using the EMNIST data set, the model was trained incrementally, starting with a small set of classes. The results of the simulation show that SIGANN is able to retain previous knowledge with a gradual forgetfulness for each learning sequence. Moreover, SIGANN can detect new classes that are hidden in the data and, therefore, proceed with incremental class learning.
ARTICLE | doi:10.20944/preprints201807.0119.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Convolutional Neural Network,Single Shot Detector, Regional Convolutional Neural Network, Machine Learning, Visualization-Localization
Online: 6 July 2018 (14:38:52 CEST)
The emerging use of visualization techniques in pathology and microbiol- ogy has been accelerated by machine learning (ML) approaches towards image preprocessing, classification, and feature extraction in an increasingly complex series of datasets. Modern Convolutional Neural Network (CNN) architectures have developed into an umbrella of vast image reinforcement and recognition methods, including a combined classification-localization of single/multi-object featured images. As a subtype neural network, CNN cre- ates a rapid order of complexity by initially detecting borderlines, edges, and colours in images for dataset construction, eventually capable in mapping intricate objects and conformities. This paper investigates the disparities between Tensorflow object detection APIs, exclusively, Single Shot Detector (SSD) Mobilenet V1 and the Faster RCNN Inception V2 model, to sample computational drawbacks in accuracy-precision vs. real time visualization capabilities. The situation of rapid ML medical image analysis is theoretically framed in regions with limited access to pathology and disease prevention departments (e.g. 3rd world and impoverished countries). Dark field mi- croscopy datasets of an initial 62 XML-JPG annotated training files were processed under Malaria and Syphilis classes. Model trainings were halted as soon as loss values were regularized and converged.
ARTICLE | doi:10.20944/preprints202108.0272.v1
Subject: Engineering, Industrial And Manufacturing Engineering Keywords: Remaining Useful Life; Deep Neural Network; Convolutional Neural Network; Genetic Optimization; Neural Network Optimization; Support Vector Regression; Depth Maps; Normal Maps; 3D Point Clouds.
Online: 12 August 2021 (10:40:23 CEST)
In the current industrial landscape, increasingly pervaded by technological innovations, the adoption of optimized strategies for asset management is becoming a critical key success factor. Among the various strategies available, the “Prognostics and Health Management” strategy is able to support maintenance management decisions more accurately, through continuous monitoring of equipment health and “Remaining Useful Life” forecasting. In the present study, Convolutional Neural Network-based Deep Neural Network techniques are investigated for the Remaining Useful Life prediction of a punch tool, whose degradation is caused by working surface deformations during the machining process. Surface deformation is determined using a 3D scanning sensor capable of returning point clouds with micrometric accuracy during the operation of the punching machine, avoiding both downtime and human intervention. The 3D point clouds thus obtained are transformed into bidimensional image-type maps, i.e., maps of depths and normal vectors, to fully exploit the potential of convolutional neural networks for extracting features. Such maps are then processed by comparing 15 genetically optimized architectures with the transfer learning of 19 pre-trained models, using a classic machine learning approach, i.e., Support Vector Regression, as a benchmark. The achieved results clearly show that, in this specific case, optimized architectures provide performance far superior (MAPE=0.058) to that of transfer learning which, instead, remains at a lower or slightly higher level (MAPE=0.416) than Support Vector Regression (MAPE=0.857).
ARTICLE | doi:10.20944/preprints202010.0613.v1
Subject: Engineering, Automotive Engineering Keywords: Gas emission prediction; grey theory; RBF neural network model; improved grey RBF neural network model
Online: 29 October 2020 (13:22:44 CET)
Effectively avoiding gas accident is vital to the security of mineral manufacture, and the coal mine gas accident is often caused by gas concentration overrun. The prediction accuracy of gas emission quantity in coal mine is the key to solve this problem. To maintain concentration of gas in a secure range，grey theory and neural network model increasingly diffusely used in forecasting gas emission quantity in coal mine critically. Nevertheless, the limitation of the grey neural network model is that researchers merely bonded the conventional neural network and grey theory. To enhance accuracy of prediction, a modified grey GM(1,1) and RBF neural network model is proposed combined amended grey GM(1,1) model and RBF neural network model. Then the proposed model was put into simulation experiment which is built based on Matlab software. Ultimately, conclusion of the simulation experiment verified that the modified grey GM(1,1) and RBF neural network model not only boosts the precision of prediction, but also restricts relative error in a minimum range. This showed that the modified grey GM(1,1) and RBF neural network model achieves effectiveness in precision of prediction much better than grey GM(1,1) model and RBF neural network model.
ARTICLE | doi:10.20944/preprints202302.0275.v2
Subject: Computer Science And Mathematics, Computer Science Keywords: Neural Network, Uncertainties , GMM, Density approximation
Online: 16 March 2023 (02:21:19 CET)
While modeling data in reality, uncertainties in both data (aleatoric) and model (epistemic) are not necessarily Gaussian. The uncertainties could appear with multimodality or any other particular forms which need to be captured. Gaussian mixture models (GMMs) are powerful tools that allow us to capture those features from a unknown density with a peculiar shape. Inspired by Fourier expansion, we propose a GMM model structure to decompose arbitrary unknown densities and prove the property of convergence. A simple learning method is introduced to learn GMMs through sampled datasets. We applied GMMs as well as our learning method to two classic neural network applications. The first one is learning encoder output of autoencoder as GMMs for handwritten digits images generation. The another one is learning gram matrices as GMMs for style-transfer algorithm. Comparing with the classic Expectation maximization (EM) algorithm, our method does not involve complex formulations and is capable of achieving applicable accuracy.
ARTICLE | doi:10.20944/preprints202209.0222.v1
Subject: Chemistry And Materials Science, Biomaterials Keywords: PEDOT:PSS; Neural recording; Immune response; BMI
Online: 15 September 2022 (08:04:38 CEST)
One of the significant challenges today in the brain-machine interface using invasive methods is the stability of the chronic record. In recent years, polymer-based electrodes have gained notoriety for achieving mechanical strength values close to that of brain tissue, promoting a lower immune response to the implant. In this work, we fabricated fully polymeric electrodes based on PEDOT:PSS for neural recording in Wistar rats. We characterized the electrical properties and both in-vitro and in-vivo functionality of the electrodes. Also, we employed histological processing and microscopical visualization to evaluate tecidual immune response in 7, 14, and 21 days post-implant days. Electrodes with 400-micrometer channels showed a 12dB signal-to-noise ratio. Local field potentials were characterized under two conditions: anesthetized and free-moving. There was a proliferation of microglia to the tissue-electrode interface in the first days, with a decrease after 14 days. Astrocytes also migrated to the interface, but there was no continuous recruitment of these cells in the tissue, showing inflammatory stability at 21 days. The signal was not affected by this inflammatory action, demonstrating that fully polymeric electrodes can be an alternative to prolong the valuable time of neural recordings.
ARTICLE | doi:10.20944/preprints202209.0212.v1
Subject: Computer Science And Mathematics, Computer Vision And Graphics Keywords: Graph Neural Network; Recommendation; Social Relationship
Online: 14 September 2022 (16:09:15 CEST)
There is a considerable amount of research in online social networks, most of which focuses on the structural analysis of social graphs. The interpersonal relationships of social networks, especially friend circle, can solve the cold start and sparsity problems, and through the relationship between social networks can effectively recommend users' favorite items (items), such as music , videos, brands/products, preferred tags, location, services, etc. User relationships in social networks are diverse and there are many different perspectives on different social networks. Associations among users can form multi-layered composite networks, and multi-layered social networks present new challenges and opportunities. Different relationships can influence users' preferences to different degrees, which in turn affects their behavior. Therefore, fusing multiple social networks is an effective way to improve recommendation. Although some studies have started to address multiple social network recommendations, simple linear superposition cannot reflect the coupling and nonlinear association between multiple social networks. In this paper, we propose a graph neural network recommendation model under social relationships based on this background. We first propose to compute the 2nd order collaborative signals and their intensities directly from the neighboring matrix for updating the node embedding of the graph convolution layer. Secondly by embedding historical evaluations, various social networks constituting different dimensions, the attention integration of user preferences by different social networks is achieved, and its effectiveness and scalability are demonstrated in theoretical derivation and experimental validation. The theoretical derivation and experimental validation demonstrate its effectiveness and scalability.
ARTICLE | doi:10.20944/preprints202205.0313.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: Failure Prediction; Asynchronous motor; Neural Network
Online: 24 May 2022 (03:37:35 CEST)
Three-phase motors are commonly adopted in several industrial contexts and their failures can result in costly downtime causing undesired service outages; this way, motor diagnostics is an issue that assumes great importance. To prevent their failures and timely face the considered service outages, a non-invasive method to identify electrical and mechanical faults in three-phase asynchronous electric motors is proposed in the paper. In particular, a measurement strategy along with a machine learning algorithm based on Artificial Neural Network is exploited to properly classify failures. In particular, digitized current samples of each motor phase are first processed by means of FFT and PSD in order to estimate the associated spectrum. Suitable features (in terms of frequency and amplitude of the spectral components) are then singled out to either train or feed a neural network acting as a classifier. The method is preliminary validated on a set of 28 electric motors, and its performance is compared with common state-of-art machine learning techniques. The obtained results show that the proposed methodology is able to reach accuracy levels greater than 98\% in identifying anomalous conditions of three-phase asynchronous motors.
ARTICLE | doi:10.20944/preprints202104.0523.v1
Subject: Computer Science And Mathematics, Data Structures, Algorithms And Complexity Keywords: Backpropagation Algorithm; Kalman Filter; Neural Networks
Online: 20 April 2021 (08:49:55 CEST)
This work describes and compares the backpropagation algorithm with the Extended Kalman filter, a second-order training method which can be applied to the problem of learning neural network parameters and is known to converge in only a few iterations. The algorithms are compared with respect to their effectiveness and speed of convergence using simulated data for both, a regression and a classification task.
ARTICLE | doi:10.20944/preprints202004.0271.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: Keratoconus; smartphone; cornea; convolutional neural network
Online: 16 April 2020 (12:38:42 CEST)
Nowadays smartphone utilization for disease diagnosis and remote health care applications has become promising due to their ubiquity. Here, a novel convolutional neural network method for detecting keratoconus that is wholly implemented on a smartphone is proposed. The proposed method provides accurate detection of over 72.9% for all stages of keratoconus. Preliminary results indicate 90%, 83%, 64% and 52% detection rate for severe, advanced, moderate and mild stages of disease, respectively.
Subject: Computer Science And Mathematics, Computer Science Keywords: depthwise; dilated; neural network; network complexity
Online: 14 September 2019 (12:40:26 CEST)
It is important to reduce the computation complexity while maintaining the accuracy of convolution neural networks. We deem it is possible to further reduce the network complexity while ensuring the accuracy. In this paper, we propose a novel feature extraction network called DSRNet which is lightweight but effective. DSRNet follows the basic ideas of stacking modules and short connection, introduces Depthwise Separable convolution and utilizes the Dilated convolution. The proposed network has fewer parameters and achieves outstanding speed. We conducted comprehensive experiments on CIFAR10, CIFAR100 and STL10 datasets, and the results showed the DSRNet has great performance improvement in terms of accuracy and speed.
ARTICLE | doi:10.20944/preprints201811.0583.v1
Subject: Computer Science And Mathematics, Data Structures, Algorithms And Complexity Keywords: Station logo; Convolutional Neural Network; Detection
Online: 26 November 2018 (10:57:17 CET)
The Station logo is a way for a TV station to claim copyright, which can realize the analysis and understanding of the video by the identification of the station logo, so as to ensure that the broadcasted TV signal will not be illegally interfered. In this paper, we design a station logo detection method based on Convolutional Neural Network by the characteristics of the station, such as small scale-to-height ratio change and relatively fixed position. Firstly, in order to realize the preprocessing and feature extraction of the station data, the video samples are collected, filtered, framed, labeled and processed. Then, the training sample data and the test sample data are divided proportionally to train the station detection model. Finally, the sample is tested to evaluate the effect of the training model in practice. The simulation experiments prove its validity.
ARTICLE | doi:10.20944/preprints201809.0305.v1
Subject: Engineering, Control And Systems Engineering Keywords: Neural Controller; Nonlinear control; Cylindrical Tank.
Online: 17 September 2018 (11:39:35 CEST)
Most systems that are the subject of control engineering studies have some non-linearity. An example of this is the horizontal cylindrical tank, commonly used in process industries. To deal with cases like this, several control theories have been developed over time, each one presenting better results in certain systems. This work presents an alternative for the control of nonlinear systems, without necessary modeling or previous information about the system, based on a new optimization law for the artificial neural network training in real time.
ARTICLE | doi:10.20944/preprints201805.0458.v1
Subject: Social Sciences, Behavior Sciences Keywords: action; perception; neural; guidance; information; organism
Online: 31 May 2018 (03:23:52 CEST)
A theory is presented about the information available for guiding purposeful actions by any organism, whether animal or plant, and about how the information is used in guiding actions.
ARTICLE | doi:10.20944/preprints201712.0156.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: Network Embedding; Neural Network; Relation Extraction
Online: 21 December 2017 (16:39:01 CET)
Network Embedding (NE) is an important method to learn the representations of network via a low-dimensional space. Conventional NE models focus on capturing the structure information and semantic information of vertices while neglecting such information for edges. In this work, we propose a novel NE model named BimoNet to capture both the structure and semantic information of edges. BimoNet is composed of two parts, i.e., the bi-mode embedding part and the deep neural network part. For bi-mode embedding part, the first mode named add-mode is used to express the entity-shared features of edges and the second mode named subtract-mode is employed to represent the entity-specific features of edges. These features actually reflect the semantic information. For deep neural network part, we firstly regard the edges in a network as nodes, and the vertices as links, which will not change the overall structure of the whole network. Then we take the nodes' adjacent matrix as the input of the deep neural network as it can obtain similar representations for nodes with similar structure. Afterwards, by jointly optimizing the objective function of these two parts, BimoNet could preserve both the semantic and structure information of edges. In experiments, we evaluate BimoNet on three real-world datasets and task of relation extraction, and BimoNet is demonstrated to outperform state-of-the-art baseline models consistently and significantly.
ARTICLE | doi:10.20944/preprints201708.0016.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: neural network; potts model; latching; recursion
Online: 4 August 2017 (14:20:59 CEST)
We study latching dynamics in the adaptive Potts model network, through numerical simulations with randomly and also weakly correlated patterns, and we focus on comparing its slowly and fast adapting regimes. A measure, Q, is used to quantify the quality of latching in the phase space spanned by the number of Potts states S, the number of connections per Potts unit C and the number of stored memory patterns p. We find narrow regions, or bands in phase space, where distinct pattern retrieval and duration of latching combine to yield the highest values of Q. The bands are confined by the storage capacity curve, for large p, and by the onset of finite latching, for low p. Inside the band, in the slowly adapting regime, we observe complex structured dynamics, with transitions at high crossover between correlated memory patterns; while away from the band latching transitions lose complexity in different ways: below, they are clear-cut but last so few steps as to span a transition matrix between states with few asymmetrical entries and limited entropy; while above, they tend to become random, with large entropy and bi-directional transition frequencies, but indistinguishable from noise. Extrapolating from the simulations, the band appears to scale almost quadratically in the p - S plane, and sublinearly in p - C. In the fast adapting regime the band scales similarly, and it can be made even wider and more robust, but transitions between anti-correlated patterns dominate latching dynamics. This suggest that slow and fast adaptation have to be integrated in a scenario for viable latching in a cortical system. The results for the slowly adapting regime, obtained with randomly correlated patterns, remain valid also for the case with correlated patterns, with just a simple shift in phase space.
Subject: Computer Science And Mathematics, Computer Science Keywords: Indoor Localization; Sensor Fusion; Multimodal Deep Neural Network; Multimodal Sensing; WiFi Fingerprinting; Pedestrian Dead Reckoning
Online: 13 October 2021 (12:14:39 CEST)
Many engineered approaches have been proposed over the years for solving the hard problem of performing indoor localisation using smartphone sensors. However, specialising these solutions for difficult edge cases remains challenging. Here we propose an end-to-end hybrid multimodal deep neural network localisation system, MM-Loc, relying on zero hand-engineered features, learning them automatically from data instead. This is achieved by using modality-specific neural networks to extract preliminary features from each sensing modality, which are then combined by cross-modality neural structures. We show that our choice of modality-specific neural architectures is capable of estimating the location with good accuracy independently. But for better accuracy, a multimodal neural network fusing the features of early modality-specific representations is a better proposition. Our proposed MM-Loc solution is tested on cross-modality samples characterised by different sampling rates and data representation (inertial sensors, magnetic and WiFi signals), outperforming traditional approaches for location estimation. MM-Loc elegantly trains directly from data unlike conventional indoor positioning systems, which rely on human intuition.
ARTICLE | doi:10.20944/preprints202102.0031.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: primary cilium; ciliogenesis; neural circuits formation; neural crest cells; DRG; boundary cap cells; sympathetic ganglia; PNS
Online: 1 February 2021 (13:26:06 CET)
The primary cilium plays a pivotal role during embryonic development of vertebrates. It acts as a somatic signaling hub for specific pathways, such as sonic hedgehog signaling. In humans, mutations in genes that cause dysregulation of ciliogenesis or ciliary function lead to severe developmental disorders called ciliopathies. Beyond its obvious role in early morphogenesis, growing evidence points towards an essential function of the primary cilium in neural circuit formation in the central nervous system. However, very little is known about a potential role in the formation of the peripheral nervous system. Here, we investigated the presence of the primary cilium in neural crest cells and their derivatives in the trunk of the developing chicken embryo in vivo. We found that neural crest cells, sensory neurons, and boundary cap cells all bear a primary cilium during key stages of early peripheral nervous system formation. Moreover, we described differences in the ciliation of neuronal cultures of different populations from the peripheral and central nervous system. Our results offer a framework for further in vivo and in vitro investigations on specific roles that the primary cilium might play during peripheral nervous system formation.
Subject: Engineering, Bioengineering Keywords: AI; deep-learning; neural-networks; graph neural-networks; cheminformatics; molecular property; machine-learning; computational chemistry; lipophilicity; solubility
Online: 1 October 2021 (14:29:01 CEST)
The accurate prediction of molecular properties such as lipophilicity and aqueous solubility are of great importance and pose challenges in several stages of the drug discovery pipeline. Machine learning methods like graph-based neural networks (GNNs) have shown exceptionally good performance in predicting these properties. In this work, we introduce a novel GNN architecture, called directed edge graph isomorphism network (D-GIN). It is composed of two distinct sub-architectures (D-MPNN, GIN) and achieves an improvement in accuracy over its sub-architectures employing various learning, and featurization strategies. We argue that combining models with different key aspects help make graph neural networks deeper and simultaneously increase their predictive power. Furthermore, we address current limitations in assessment of deep-learning models, namely, comparison of single training run performance metrics, and offer a more robust solution.
ARTICLE | doi:10.20944/preprints202007.0656.v1
Subject: Medicine And Pharmacology, Epidemiology And Infectious Diseases Keywords: COVID-19 infection; Chest X-ray image; generalized regression neural network; probabilistic neural network and detection accuracy
Online: 27 July 2020 (00:52:49 CEST)
Corona virus disease (COVID-19) has infected over more than 10 million people around the globe and killed at least 500K worldwide by the end of June 2020. As this disease continues to evolve and scientists and researchers around the world now trying to find out the way to combat this disease in most effective way. Chest X-rays are widely available modality for immediate care in diagnosing COVID-19. Precise detection and diagnosis of COVID-19 from these chest X-rays would be practical for the current situation. This paper proposes one shot cluster based approach for the accurate detection of COVID-19 chest x-rays. The main objective of one shot learning (OSL) is to mimic the way humans learn in order to make classification or prediction on a wide range of similar but novel problems. The core constraint of this type of task is that the algorithm should decide on the class of a test instance after seeing just one test example. For this purpose we have experimented with widely known Generalized Regression and Probabilistic Neural Networks. Experiments conducted with publicly available chest x-ray images demonstrate that the method can detect COVID-19 accurately with high precision. The obtained results have outperformed many of the convolutional neural network based existing methods proposed in the literature.
REVIEW | doi:10.20944/preprints202005.0460.v1
Subject: Biology And Life Sciences, Neuroscience And Neurology Keywords: extracellular vesicles; exosomes; neural tissue repair; neuroregeneration; non-human primates; hydrogels; neural tissue engineering; stroke; cortical injury
Online: 28 May 2020 (13:08:32 CEST)
Neural tissue engineering, nanotechnology and neuroregeneration are diverse biomedical disciplines that have been working together in recent decades to solve the complex problems linked to central nervous system (CNS) repair. It is known that the CNS demonstrates a very limited regenerative capacity because of a microenvironment that impedes effective regenerative processes, making development of CNS therapeutics challenging. Given the high prevalence of CNS conditions such as stroke that damage the brain and place a severe burden on afflicted individuals and on society, it is of utmost significance to explore the optimum methodologies for finding treatments that could be applied to humans for restoration of function to pre-injury levels. Extracellular vesicles (EVs), also known as exosomes, when derived from mesenchymal stem cells, are one of the most promising approaches that have been attempted thus far, as EVs deliver factors that stimulate recovery by acting at the nanoscale level on intercellular communication while avoiding the risks linked to stem cell transplantation. At the same time, advances in tissue engineering and regenerative medicine have offered the potential of using hydrogels as bio-scaffolds in order to provide the stroma required for neural repair to occur, as well as the release of biomolecules facilitating or inducing the reparative processes. This review introduces a novel experimental hypothesis regarding the benefits that could be offered if EVs were to be combined with biocompatible injectable hydrogels. The rationale behind this hypothesis is presented, analyzing how a hydrogel might prolong the retention of EVs and maximize the localized benefit to the brain. This sustained delivery of EVs would be coupled with essential guidance cues and structural support from the hydrogel until neural tissue remodeling and regeneration occur. Finally, the importance of including non-human primate (NHP) models in the clinical translation pipeline, as well as the added benefit of multi-modal neuroimage analysis to establish non-invasive, in vivo, quantifiable imaging-based biomarkers for CNS repair are discussed, aiming for more effective and safe clinical translation of such regenerative therapies to humans.
ARTICLE | doi:10.20944/preprints201902.0233.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: deep neural network architectures; supervised learning; unsupervised learning; testing neural networks; applications of deep learning; evolutionary computation
Online: 26 February 2019 (04:02:00 CET)
Deep learning has taken over - both in problems beyond the realm of traditional, hand-crafted machine learning paradigms as well as in capturing the imagination of the practitioner sitting on top of petabytes of data. While the public perception about the efficacy of deep neural architectures in complex pattern recognition tasks grows, sequentially up-to-date primers on the current state of affairs must follow. In this review, we seek to present a refresher of the many different stacked, connectionist networks that make up the deep learning architectures followed by automatic architecture optimization protocols using multi-agent approaches. Further, since guaranteeing system uptime is fast becoming an indispensable asset across multiple industrial modalities, we include an investigative section on testing neural networks for fault detection and subsequent mitigation. This is followed by an exploratory survey of several application areas where deep learning has emerged as a game-changing technology - be it anomalous behavior detection in financial applications or financial time-series forecasting, predictive and prescriptive analytics, medical imaging, natural language processing or power systems research. The thrust of this review is on outlining emerging areas of application-oriented research within the deep learning community as well as to provide a handy reference to researchers seeking to embrace deep learning in their work for what it is: statistical pattern recognizers with unparalleled hierarchical structure learning capacity with the ability to scale with information.
ARTICLE | doi:10.20944/preprints202305.2163.v1
Subject: Engineering, Bioengineering Keywords: chickpea; convolutional neural network; transfer learning; classification
Online: 31 May 2023 (03:32:49 CEST)
Chickpea is one of the most widely consumed pulses globally because of its high protein content. The morphological features of chickpea seed, such as colour, texture, are observable and play a major role in classifying different chickpea varieties. This process is often carried out by human experts, and is time-consuming, inaccurate, and expensive. The objective of the study was to design an automated chickpea classifier using an RGB colour image-based model by considering the morphological features of chickpea seed. As part of the data acquisition process, five hundred and fifty images were collected per variety for four varieties of chickpea (CDC-Alma, CDC-Consul, CDC-Cory, and CDC-Orion) using an industrial RGB camera and a mobile phone camera. Three CNN-based models such as NasNet-A (mobile), MobileNetV3 (small), and EfficientNetB0 were evaluated using a transfer learning-based approach. The classification accuracy was 97%, 99%, and 98% for NasNet-A (mobile), MobileNetV3 (small), and EfficientNetB0 models, respectively. The MobileNetV3 model was used for further deployment on an Android mobile and Raspberry Pi 4 devices based on its higher accuracy and light-weight architecture. The classification accuracy for the four chickpea varieties was 100% while the MobileNetV3 model was deployed on both Android mobile and Raspberry Pi 4 platforms.
ARTICLE | doi:10.20944/preprints202305.1843.v1
Subject: Physical Sciences, Optics And Photonics Keywords: Uncertainty; Neural Networks; Bayesian Inversion; Remote Sensing
Online: 26 May 2023 (04:22:05 CEST)
The Ocean Color - Simultaneous Marine and Aerosol Retrieval Tool (OC-SMART) is a robust data processing platform that supports a large array of multi-spectral and hyper-spectral sensors. It provides accurate aerosol optical depths and remote sensing reflectances (Rrs estimates) that can be used to generate products such as absorption coefficients due to phytoplankton and detritus/Gelbstoff as well as backscattering coefficients due to particulate matter. The OC-SMART platform yields improved performance in complex environments by utilizing scientific machine learning (SciML) in conjunction with comprehensive radiative transfer computations. This paper expands the capability of OC-SMART by quantifying uncertainties in ocean color retrievals. Bayesian inversion is used to relate measured top of atmosphere radiances and a priori data to estimate posterior probability density functions and associated uncertainties. A framework of the methodology and implementation strategy is presented and uncertainty estimates for Rrs retrievals are provided to demonstrate the approach by applying it to MODIS, OLCI Sentinel-3, and VIIRS sensor data.
ARTICLE | doi:10.20944/preprints202305.0797.v1
Subject: Computer Science And Mathematics, Data Structures, Algorithms And Complexity Keywords: link prediction; graph neural network; graph embedding
Online: 11 May 2023 (05:24:00 CEST)
Link prediction is to complete the missing links in the network or to predict the generation of new links according to the current network structure information, which is very important for mining and analyzing the evolution of the network such for construction and analysis of logical architecture in 6G network. Link prediction algorithms based on node similarity need predefined similarity functions, which is highly hypothetical and only applies to specific network structures without generality. To solve this problem, this paper proposes a link prediction algorithm based on the subgraph of the target node pair. In order to automatically learn the graph structure characteristics, the algorithm firstly extracts the h-hop subgraph of the target node pair, and then predicts whether the target node pair will be linked according to the subgraph. Experiments on seven real data sets show that the link prediction algorithm based on target node pair subgraph is suitable for various network structures and superior to other link prediction algorithms.
ARTICLE | doi:10.20944/preprints202305.0067.v1
Subject: Computer Science And Mathematics, Computer Vision And Graphics Keywords: low light; image enhancement; counvolutional neural networks
Online: 2 May 2023 (07:32:56 CEST)
In this study, we explore the potential of using a straightforward neural network inspired by the retina model to efficiently restore low-light images. The retina model imitates the neurophysiological principles and dynamics of various optical neurons. Our proposed neural network model reduces the computational overhead compared to traditional signal-processing models while achieving results similar to complex deep learning models from a subjective perceptual perspective. By directly simulating retinal neuron functionalities with neural networks, we not only avoid manual parameter optimization but also lay the groundwork for constructing artificial versions of specific neurobiological organizations.
ARTICLE | doi:10.20944/preprints202303.0345.v2
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: neural architecture search; machine learning; computer vision
Online: 29 March 2023 (02:14:14 CEST)
Existing one-shot neural architecture search (NAS) methods have to conduct a search over a giant super-net, which leads to the huge computational cost. To reduce such cost, in this paper, we propose a method, called FTSO, to divide the whole architecture search into two sub-steps. Specifically, in the first step, we only search for the topology, and in the second step, we search for the operators. FTSO not only reduces NAS’s search time from days to 0.68 seconds, but also significantly improves the found architecture's accuracy. Our extensive experiments on ImageNet show that within 18 seconds, FTSO can achieve a 76.4% testing accuracy, 1.5% higher than the SOTA, PC-DARTS. In addition, FTSO can reach a 97.77% testing accuracy, 0.27% higher than the SOTA, with nearly 100% (99.8%) search time saved, when searching on CIFAR10.
REVIEW | doi:10.20944/preprints202303.0159.v1
Subject: Biology And Life Sciences, Anatomy And Physiology Keywords: neural crest; mesectoderm; embryology; evolution; forebrain development
Online: 9 March 2023 (01:25:14 CET)
The neural crest, a unique cell population originating from the primitive neural field, has a multi-systemic and structural contribution to vertebrate development. At the cephalic level, the neural crest generates most of the skeletal tissues encasing the developing forebrain and provides the prosencephalon with functional vasculature and meninges. Over the last decade, we have demonstrated that the cephalic neural crest (CNC) exerts an autonomous and prominent control on forebrain and sense organs development. The present paper reviews the primary mechanisms by which CNC can orchestrate vertebrate encephalization. Demonstrating the role of the CNC as an exogenous source of patterning for the forebrain provides a novel conceptual framework with profound implications for understanding neurodevelopment. From a biomedical standpoint, these data suggest that the spectrum of neurocristopathies is broader than expected and that some neurological disorders may stem from CNC dysfunctions.
ARTICLE | doi:10.20944/preprints202212.0478.v1
Subject: Computer Science And Mathematics, Information Systems Keywords: Datasets, Neural Networks, Hand Detection, Text Tagging
Online: 26 December 2022 (07:30:24 CET)
American Sign Language is a popular language for deaf individuals. Communication is made easier for these people through sign language. However, in a digital era like today, there is a need for these people to be able to communicate online, and even get help from technology to communicate in person with non sign language speakers. This research will present a program able to translate American sign language to plain English. This study aims to use the OpenCV library to recognize hand signals, also a trained model to identify images so that the program can then translate them to words and letters. The program uses a data set of over 2000 images which will be in this case the largest data set available. With over 90\% of accuracy it results in a basic computer program with the largest data set available that would make possible for users to communicate with a wide variety of words and expressions.
ARTICLE | doi:10.20944/preprints202211.0226.v1
Subject: Computer Science And Mathematics, Analysis Keywords: deep learning; convolutional neural networks; remote sensing
Online: 14 November 2022 (01:20:07 CET)
Deep Learning is an extremely important research topic in Earth Observation. Current use-cases range from semantic image segmentation, object detection to more common problems found in computer vision such as object identification. Earth Observation is an excellent source for different types of problems and data for Machine Learning in general and Deep Learning in particular. It can be argued that both Earth Observation and Deep Learning as fields of research will benefit greatly from this recent trend of research. In this paper we take several state of the art Deep Learning network topologies and provide a detailed analysis of their performance for semantic image segmentation for building footprint detection. The dataset used is comprised of high resolution images depicting urban scenes. We focused on single model performance on simple RGB images. In most situations several methods have been applied to increase the accuracy of prediction when using deep learning such as ensembling, alternating between optimisers during training and using pretrained weights to bootstrap new models. These methods although effective, are not indicative of single model performance. Instead, in this paper, we present different topology variations of these state of the art topologies and study how these variations effect both training convergence and out of sample, single model, performance.
ARTICLE | doi:10.20944/preprints202210.0186.v3
Subject: Biology And Life Sciences, Biophysics Keywords: neural; brain; structural intelligence; cell expression; evolution
Online: 4 November 2022 (09:43:59 CET)
This concept paper gives a narrative about intelligence from insects to the human brain, showing where evolution may have been influenced by the structures in these simpler organisms. The ideas also come from the author's own cognitive model, where a number of algorithms have been developed over time and the precursor structures should be codable to some level. Through developing and trying to implement the design, ideas like separating the data from the function have become architecturally appropriate and there have been several opportunities to make the system more orthogonal. Similarly for the human brain, neural structures may work in-sync with the neural functions, or may be slightly separate from them. Each section discusses one of the neural assemblies with a potential functional result, that cover ideas such as timing or scheduling, structural intelligence and neural binding. Another aspect of self-representation or expression is interesting and may help the brain to realise higher-level functionality based on these lower-level processes.
ARTICLE | doi:10.20944/preprints202210.0437.v1
Subject: Social Sciences, Behavior Sciences Keywords: methamphetamine; mice; neural oscillations; gamma power; sensitisation
Online: 28 October 2022 (02:34:58 CEST)
Dysregulation of high-frequency neuronal oscillations has been implicated in the pathophysiology of schizophrenia. Chronic methamphetamine (METH) use can induce psychosis similar to paranoid schizophrenia. The current study in mice aimed to determine the effect of chronic METH treatment on ongoing and evoked neuronal oscillations. C57BL/6 mice were treated with METH or vehicle control for three weeks and implanted with extradural recording electrodes. Two weeks after the last METH injection, mice underwent three EEG recording sessions to measure ongoing and auditory-evoked gamma and beta oscillatory power in response to an acute challenge with METH (2mg/kg), the NMDA receptor antagonist MK-801 (0.3mg/kg), or saline control. A separate group of mice pretreated with METH showed significantly greater locomotor hyperactivity to an acute METH challenge, confirming long-term sensitisation. Chronic METH did not affect ongoing or evoked gamma or beta power. Acute MK-801 challenge reduced ongoing beta power whereas acute METH challenge significantly increased ongoing gamma power. Both MK-801 and METH challenge suppressed evoked gamma power. Chronic METH treatment did not modulate these acute drug effects. There were minor effects of chronic METH and acute METH and MK-801 on selected components of event-related potential (ERP) waves. In conclusion, chronic METH treatment did not exert neuroplastic effects on the regulation of cortical gamma oscillations in a manner consistent with schizophrenia, despite causing behavioural sensitisation.
ARTICLE | doi:10.20944/preprints202209.0226.v1
Subject: Computer Science And Mathematics, Data Structures, Algorithms And Complexity Keywords: Graph Neural Networks, Non-Euclidean Data, Bioinformation
Online: 15 September 2022 (08:49:51 CEST)
With the development of data science, more and more machine learning technologies have been designed to solve complicated and challenging real-world tasks containing a large volume of data. And many significant real-world datasets contain data in the form of networks or graphs. Graph Neural Networks is one of the powerful machine learning tools, which could provide a perfect solution to processing a large amount of non-Euclidean data. And because most bio information data in bioinformatics is in the non-Euclidean domain, Graph Neural Networks could then directly be applied to solve problems in bioinformatics. Much research has been done in the field of GNN, and there are also some surveys related to GNN and its applications. However, there has been little research focusing on GNN in bioinformatics, and we think in the future we could better utilize GNN in the field of biology, so we would like to write a literature review to help take a comprehensive look at GNN and their applications in the field of bioinformatics. In this paper, we would first introduce SOTA models in Graph Neural Networks, and second, introduce their applications in bio information. And then we would provide future directions of Graph Neural Networks in bioinformatics.
ARTICLE | doi:10.20944/preprints202111.0186.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: Explainable AI; Convolutional Neural Network; Network Compression
Online: 9 November 2021 (15:03:27 CET)
Model understanding is critical in many domains, particularly those involved in high-stakes decisions, i.e., medicine, criminal justice, and autonomous driving. Explainable AI (XAI) methods are essential for working with black-box models such as Convolutional Neural Networks. This paper evaluates the traffic sign classifier of Deep Neural Network (DNN) from the Programmable Systems for Intelligence in Automobiles (PRYSTINE) project for explainability. The results of explanations were further used for the CNN PRYSTINE classifier vague kernels` compression. After all, the precision of the classifier was evaluated in different pruning scenarios. The proposed classifier performance methodology was realised by creating the original traffic sign and traffic light classification and explanation code. First, the status of the kernels of the network was evaluated for explainability. For this task, the post-hoc, local, meaningful perturbation-based forward explainable method was integrated into the model to evaluate each kernel status of the network. This method enabled distinguishing high and low-impact kernels in the CNN. Second, the vague kernels of the classifier of the last layer before the fully connected layer were excluded by withdrawing them from the network. Third, the network's precision was evaluated in different kernel compression levels. It is shown that by using the XAI approach for network kernel compression, the pruning of 5% of kernels leads only to a 1% loss in traffic sign and traffic light classification precision. The proposed methodology is crucial where execution time and processing capacity prevail.
ARTICLE | doi:10.20944/preprints202109.0223.v1
Subject: Business, Economics And Management, Accounting And Taxation Keywords: Accounting Distortions; Financial Statements; Neural Network; Accounting
Online: 13 September 2021 (16:18:55 CEST)
Distortion of financial statements is recognized as one of the most important issues in the field of accounting and auditing, which is also one of the most common issues today. In this regard, the present research was conducted, in which stock exchange information was used to investigate, predict, and model accounting distortions. For this purpose, financial performance, non-financial metrics, market-based metrics and commitment, or selection items were reviewed over a 6-year period. For collecting data of distorting companies, database of the Society of Certified Public Accountants in Iran was used and the information was analyzed using data mining methods (decision tree, neural networks, and Bayesian method). The results showed that analysis of financial statements҆ information has a high accuracy in determining and identifying the distorted financial statements. Using this information, it is possible to get better acquainted with the methods of document distortion and to take necessary measures in order to control and prevent administrative violations at national and international levels. Given frequent occurrence of these violations, artificial intelligence models can be used to identify these papers.
HYPOTHESIS | doi:10.20944/preprints202102.0349.v4
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Learning (artificial intelligence); Neural networks; Activity recognition
Online: 7 June 2021 (13:04:11 CEST)
Over the past decade, recognition of human activities (HAR) has become a vibrant field of research, in particular, the spread in our everyday lives of electronics such as mobile phones, smart cell phones, and video cameras. Furthermore, the advancement in the field of deep methodologies and other paradigms have enabled scientists to enable HAR in many areas, consisting of activities in fitness and wellness. For instance, HAR is one of many resorts to support older people through day-to-day activities to support their cognition and physicality. This study is centered on the key aspects deep learning plays in the development of HAR applications. Although numerous HAR examination studies were carried out previously, there have been no overall studies on this subject, in all the earlier studies there were only specific HAR-related subjects. A detailed review covering all the main subjects in this area is therefore essential. This study discusses the latest developments and works in HAR. It separates the methods and the advantages and disadvantages of each method group. This paper finally addresses many problems in the current HAR subject and provides recommendations for potential study.
Subject: Computer Science And Mathematics, Algebra And Number Theory Keywords: Neural Networks; Machine Learning; Bootstrap; Resampling; Algorithms
Online: 22 March 2021 (16:09:04 CET)
Neural networks present the characteristics that the results are strongly dependent on the training data, the weight initialisation, and the hyper-parameters chosen. The determination of the distribution of a statistical estimator, as the Mean Squared Error (MSE) or the accuracy, is fundamental to evaluate the performance of a neural network model (NNM). For many machine learning models, as linear regression, it is possible to analytically obtain information as variance or confidence intervals on the results. Neural networks present the difficulty of being not analytically tractable due to their complexity. Therefore, it is impossible to easily estimate distributions of statistical estimators. When estimating the global performance of an NNM by estimating the MSE in a regression problem, for example, it is important to know the variance of the MSE. Bootstrap is one of the most important resampling techniques to estimate averages and variances, between other properties, of statistical estimators. In this tutorial, the application of two resampling (including bootstrap) techniques to the evaluation of neural networks’ performance is explained from both a theoretical and practical point of view. Pseudo-code of the algorithms is provided to facilitate their implementation. Computational aspects, as the training time, are discussed since resampling techniques always require to run simulations many thousands of times and, therefore, are computationally intensive. A specific version of the bootstrap algorithm is presented that allows the estimation of the distribution of a statistical estimator when dealing with an NNM in a computationally effective way. Finally, algorithms are compared on synthetically generated data to demonstrate their performance.
ARTICLE | doi:10.20944/preprints202103.0220.v1
Subject: Environmental And Earth Sciences, Atmospheric Science And Meteorology Keywords: Convolutional Neural Network; Deep Learning; Environmental Monitoring
Online: 8 March 2021 (13:37:58 CET)
Accurately mapping individual tree species in densely forested environments is crucial to forest inventory. When considering only RGB images, this is a challenging task for many automatic photogrammetry processes. The main reason for that is the spectral similarity between species in RGB scenes, which can be a hindrance for most automatic methods. State-of-the-art deep learning methods could be capable of identifying tree species with an attractive cost, accuracy, and computational load in RGB images. This paper presents a deep learning-based approach to detect an important multi-use species of palm trees (Mauritia flexuosa; i.e., Buriti) on aerial RGB imagery. In South-America, this palm tree is essential for many indigenous and local communities because of its characteristics. The species is also a valuable indicator of water resources, which comes as a benefit for mapping its location. The method is based on a Convolutional Neural Network (CNN) to identify and geolocate singular tree species in a high-complexity forest environment, and considers the likelihood of every pixel in the image to be recognized as a possible tree by implementing a confidence map feature extraction. This study compares the performance of the proposed method against state-of-the-art object detection networks. For this, images from a dataset composed of 1,394 airborne scenes, where 5,334 palm-trees were manually labeled, were used. The results returned a mean absolute error (MAE) of 0.75 trees and an F1-measure of 86.9%. These results are better than both Faster R-CNN and RetinaNet considering equal experiment conditions. The proposed network provided fast solutions to detect the palm trees, with a delivered image detection of 0.073 seconds and a standard deviation of 0.002 using the GPU. In conclusion, the method presented is efficient to deal with a high-density forest scenario and can accurately map the location of single species like the M flexuosa palm tree and may be useful for future frameworks.
ARTICLE | doi:10.20944/preprints202011.0511.v2
Subject: Computer Science And Mathematics, Computer Vision And Graphics Keywords: rumor detection; graph neural network; artificial intelligence
Online: 22 December 2020 (14:23:02 CET)
Identifying fake news on the media has been an important issue. This is especially true considering the wide spread of rumors on the popular social networks such as Twitter. Various kinds of techniques have been proposed for automatic rumor detection. In this work, we study the application of graph neural networks for rumor classification at a lower level, instead of applying existing neural network architectures to detect rumors. The responses to true rumors and false rumors display distinct characteristics. This suggests that it is essential to capture such interactions in an effective manner for a deep learning network to achieve better rumor detection performance. To this end we present a simplified aggregation graph neural network architecture. Experiments on publicly available Twitter datasets demonstrate that the proposed network has performance on a par with or even better than that of state-of-the-art graph convolutional networks, while significantly reducing the computational complexity.
ARTICLE | doi:10.20944/preprints202008.0637.v1
Subject: Medicine And Pharmacology, Oncology And Oncogenics Keywords: glioblastoma; neural stem cells; replicative senescence; metastasis
Online: 28 August 2020 (11:33:23 CEST)
Due to its aggressive and invasive nature glioblastoma (GBM), the most common and aggressive primary brain tumour in adults, remains almost invariably lethal. Significant advances in the last several years have elucidated much of the molecular and genetic complexities of GBM. However, GBM exhibits a vast genetic variation and a wide diversity of phenotypes that has complicated the development of effective therapeutic strategies. This complex pathogenesis makes it necessary the development of experimental models that could be used to further understand the disease, and also to provide a more realistic testing ground for potential therapies. In this report, we describe the process of transformation of primary mouse embryo astrocytes into immortalized cultures with neural stem cell characteristics, that are able to generate of GBM when injected in the brain of C57BL/6 mice, or heterotopic tumours when injected iv. Overall, our results show that oncogenic transformation is a fate for NSC if cultured for long periods in vitro. In addition, since no additional hit is necessary to induce the oncogenic transformation, our model may be used to investigate the pathogenesis of gliomagenesis and to test the effectiveness of different drugs throughout the natural history of GBM.
HYPOTHESIS | doi:10.20944/preprints202008.0195.v1
Subject: Social Sciences, Behavior Sciences Keywords: Information Circulation; Perception; Memory; Consciousness; Neural Circuits
Online: 7 August 2020 (11:47:54 CEST)
Long-Term Potentiation(LTP) and Long-Term Depression(LTD) are two major forms of synaptic plasticity, which are also two well-know functional and unit activities involved in high advanced central neural system(CNS) activities, like memory. But we still know little about how the advanced CNS activities are organized in the brain and in the level of organism. Based on the current understanding and experimental evidence of neurology, we propose the term “Information Circulation” to summarize the current understandings for advanced CNS activities, and we define it as separately input neural signals finally converge in different levels of CNS and interact with each other, then neural information are circulated and processed in different levels of CNS to give out orders for next body actions. This review provides a detailed description for the functional organizations of advanced CNS activities in the term of Information Circulation. This article outlines the receiving of outside stimulation and transmission of neural information, especially transmission and procession of visual bioelectrical signals, then we described neural circuits of Information Circulation in advanced CNS activities, the corresponding specificity and dynamic properties of neural circuits, different sensation linkages, and neural synchronization for information circulation to produce consciousness in CNS. In conclusion, Information Circulation is defined as an important signature involved in advanced CNS activities.
Subject: Computer Science And Mathematics, Computer Networks And Communications Keywords: artificial neural networks; rock breakage; rock blasting
Online: 23 July 2020 (10:16:24 CEST)
Over the years, various models have been developed in the stages of the mining process that have allowed predicting and enhancing results, but it is the breakage the variable that connects all the activities of the mining process from the point of view of costs (drilling, blasting, loading, hauling, crushing and grinding). To improve this process, we come up with an idea to develop a breakage prediction model; on the basis of the main variables involved in the drilling and blasting process. For this purpose, we design a computer model based on an Artificial Neuronal Networks (ANN), built by using the most representative variables that come into play with drilling and blasting, such as: the properties of the explosives, the geomechanical parameters of the rock mass, and the design parameters of drilling-blasting. For its experimentation and validation, we have taken the data from a copper mine as reference located in the north of Chile, because of we have the dataset of that ore deposit, which is valid and reliable to evaluate prediction problems based on ANN applied to copper mines. The ANN architecture was of the supervised type, feedforward, with 3 layers and 13 neurons in the only hidden layer, trained with the input data using a dataset with the previously mentioned variables, which then were compared with the breakage results. The model was feed backed in its learning process until it becomes perfected, and is a prediction option that can be used in future blasting of ore deposits with similar characteristics using the same representative variables. Therefore, this is a valid alternative for predicting rock breakage, given that it has been experimentally validated, and has achieved moderately reliable results, providing higher correlation coefficients than traditional models, and with the additional advantage that an ANN model provides, due to its ability to learn and recognize compiled dataset patterns. In this way, using this computer model we can obtain satisfactory results that allow us to predict breakage, providing an alternative for evaluating the costs that this entails.
ARTICLE | doi:10.20944/preprints202007.0379.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Transfer Learning; Convolutional Neural Networks; Emotion Recognition
Online: 17 July 2020 (13:58:18 CEST)
The paper concludes the first research on mouth-based Emotion Recognition (ER), adopting a Transfer Learning (TL) approach. Transfer Learning results paramount for mouth-based emotion ER, because a few data sets are available, and most of them include emotional expressions simulated by actors, instead of adopting a real-world categorization. Using TL we can use fewer training data than training a whole network from scratch, thus more efficiently fine-tuning the network with emotional data and improving the convolutional neural network accuracy in the desired domain. The proposed approach aims at improving the Emotion Recognition dynamically, taking into account not only new scenarios but also modified situations with respect to the initial training phase, because the image of the mouth can be available even when the whole face is visible only in an unfavourable perspective. Typical applications include automated supervision of bedridden critical patients in an healthcare management environment, or portable applications supporting disabled users having difficulties in seeing or recognizing facial emotions. This work takes advantage from previous preliminary works on mouth-based emotion recognition using CNN deep-learning, and has the further benefit of testing and comparing a set of networks on large data sets for face-based emotion recognition well known in literature. The final result is not directly comparable with works on full-face ER, but valorizes the significance of mouth in emotion recognition, obtaining consistent performances on the visual emotion recognition domain.
Subject: Computer Science And Mathematics, Computer Science Keywords: neural network; neuron; CR-PNN; Gang transform
Online: 14 June 2020 (14:32:10 CEST)
In recent years, artificial neural networks (ANNs) have won numerous contests in pattern recognition, machine learning, and artificial intelligence. The basic unit of an ANN is to mimic neurons in the brain. Neuron in ANNs is expressed as f (wx+b) or f (wx). This structure does not consider the information processing capabilities of dendrites. However, recently, studies shown that dendrites participate in pre-calculation in the brain. Concretely, biological dendrites play a role in the pre-processing to the interaction information of input data. Therefore, it is time to perfect the neuron of the neural network. This paper, add dendrite processing section, presents a novel artificial neuron, according to our previous studies (CR-PNN or Gang transform). The dendrite processing section can be expressed as . Because I perfected the basic unit of ANNs— neuron, there are so many networks to try. This article gives the basic architecture for reference in future research.
ARTICLE | doi:10.20944/preprints202004.0421.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: COVID-19; trend prediction; optimized neural network
Online: 24 April 2020 (02:57:32 CEST)
The recent worldwide outbreak of the novel corona-virus (COVID-19) opened up new challenges to the research community. Artificial intelligence (AI) driven methods can be useful to predict the parameters, risks, and effects of such an epidemic. Such predictions can be helpful to control and prevent the spread of such diseases. The main challenges of applying AI is the small volume of data and the uncertain nature. Here, we propose a shallow Long short-term memory (LSTM) based neural network to predict the risk category of a country. We have used a Bayesian optimization framework to optimized and automatically design country-specific networks. We have combined the trend data and weather data together for the prediction. The results show that the proposed pipeline outperforms against state-of-the-art methods for 170 countries data and can be a useful tool for such risk categorization. The tool can be used to predict long-duration outbreak of such an epidemic such that we can take preventive steps earlier.
ARTICLE | doi:10.20944/preprints201910.0137.v1
Subject: Computer Science And Mathematics, Mathematics Keywords: topology optimization; convolutional neural network; high-resolution
Online: 12 October 2019 (03:56:19 CEST)
Topology optimization is a pioneering design method that can provide various candidates with high mechanical properties. However, the high-resolution for the optimum structures is highly desired, normally in turn leading to computationally intractable puzzle, especially for the famous Solid Isotropic Material with Penalization (SIMP) method. In this paper, an efficient and high-resolution topology optimization method is proposed based on the Super-Resolution Convolutional Neural Network (SRCNN) technique in the framework of SIMP. The SRCNN includes four processes, i.e. refining, path extraction & representation, non-linear mapping, and reconstruction. The high computational efficiency is achieved by a pooling strategy, which can balance the number of finite element analysis (FEA) and the output mesh in optimization process. To further reduce the high computational cost of 3D topology optimization problems, a combined treatment method using 2D SRCNN is built as another speeding-up strategy. A number of typical examples justify that the high-resolution topology optimization method adopting SRCNN has excellent applicability and high efficiency for 2D and 3D problems with arbitrary boundary conditions, any design domain shape, and varied load.
ARTICLE | doi:10.20944/preprints201907.0241.v1
Subject: Physical Sciences, Astronomy And Astrophysics Keywords: recurrent neural networks, LSTM, lightcurves, simulation, variability
Online: 22 July 2019 (10:41:25 CEST)
With an explosion of data in the near future, from observatories spanning from radio to gamma-rays, we have entered the era of time domain astronomy. Historically, this field has been limited to modeling the temporal structure with time-series simulations limited to energy ranges blessed with excellent statistics as in X-rays. In addition to ever increasing volumes and variety of astronomical lightcurves, there's a plethora of different types of transients detected not only across the electromagnetic spectrum, but indeed across multiple messengers like counterparts for neutrino and gravitational wave sources. As a result, precise, fast forecasting and modeling the lightcurves or time-series will play a crucial role in both understanding the physical processes as well as coordinating multiwavelength and multimessenger campaigns. In this regard, deep learning algorithms such as recurrent neural networks (RNNs) should prove extremely powerful for forecasting as it has in several other domains. Here we test the performance of a very successful class of RNNs, the Long Short Term Memory (LSTM) algorithms with simulated lightcurves. We focus on univariate forecasting of types of lightcurves typically found in active galactic nuclei (AGN) observations. Specifically, we explore the sensitivity of training and test losses to key parameters of the LSTM network and data characteristics namely gaps and complexity measured in terms of number of Fourier components. We find that typically, the performances of LSTMs are better for pink or flicker noise type sources. The key parameters on which performance is dependent are batch size for LSTM and the gap percentage of the lightcurves. While a batch size of $10-30$ seems optimal, the most optimal test and train losses are under $10 \%$ of missing data for both periodic and random gaps in pink noise. The performance is far worse for red noise. This compromises detectability of transients. The performance gets monotonically worse for data complexity measured in terms of number of Fourier components which is especially relevant in the context of complicated quasi-periodic signals buried under noise. Thus, we show that time-series simulations are excellent guides for use of RNN-LSTMs in forecasting.
ARTICLE | doi:10.20944/preprints201904.0091.v5
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Preference learning; Multi-label ranking; Neural network; Kendall’s tau; Preference mining
Online: 19 April 2023 (07:43:17 CEST)
Equality and incomparability multi-label ranking have not been introduced to learning before. This paper proposes new native ranker neural network to address the problem of multi-label ranking including incomparable preference orders using a new activation and error functions and new architecture. Preference Neural Network PNN solves the multi-label ranking problem, where labels may have indifference preference orders or subgroups which are equally ranked. PNN is a nondeep, multiple-value neuron, single middle layer and one or more output layers network. PNN uses a novel positive smooth staircase (PSS) or smooth staircase (SS) activation function and represents preference orders and Spearman ranking correlation as objective functions. It is introduced in two types, Type A is traditional NN architecture and Type B uses expanding architecture by introducing new type of hidden neuron has multiple activation function in middle layer and duplicated output layers to reinforce the ranking by increasing the number of weights. PNN accepts single data instance as inputs and output neurons represent the number of labels and output value represents the preference value. PNN is evaluated using a new preference mining data set that contains repeated label values which have not experimented on before. SS and PS speed-up the learning and PNN outperforms five previously proposed methods for strict label ranking in terms of accurate results with high computational efficiency.
BRIEF REPORT | doi:10.20944/preprints201902.0257.v2
Subject: Engineering, Control And Systems Engineering Keywords: convolutional neural networks; pattern recognition; machine learning
Online: 12 March 2019 (10:18:12 CET)
This paper presents a study and implementation of a convolutional neural network to identify and recognize humpback whale specimens from the unique patterns of their tails. Starting from a dataset composed of images of whale tails, all the phases of the process of creation and training of a neural network are detailed – from the analysis and pre-processing of images to the elaboration of predictions, using TensorFlow and Keras frameworks. Other possible alternatives are also explained when it comes to tackling this problem and the complications that have arisen during the process of developing this paper.
ARTICLE | doi:10.20944/preprints201803.0048.v2
Subject: Biology And Life Sciences, Forestry Keywords: land-cover change; deforestation; neural network; GIS
Online: 27 December 2018 (11:42:03 CET)
Geospatial Information Systems (GIS) can provide a great environment for using machine learning algorithm for spatial data such as satellite images. Integrating this functionality with artificial intelligence algorithms for analyzing spatial data enables us to predict challenging disasters such as deforestation. Deforestation as an environmental problems has been recorded the most serious threat to environmental diversity and one of the main components of land-use change. In this paper, we investigate spatial distribution of deforestation using artificial neural networks and satellite imagery. We modeled deforestation process using various factors in determining the relationship between deforestation and environmental and socioeconomic factors. Hence, for this purpose, the proximity to roads and habitats, fragmentation of the forest, height from sea level, slope, and soil type are considered in the model. In this research, we modeled land cover changes (forests) to predict deforestation using an artificial neural network due to its significant potential for the development of nonlinear complex models. The procedure involves image registration and error correction, image classification, preparing deforestation maps, determining layers, and designing a multi-layer neural network to predict deforestation. The satellite images for this study are of a region in Hong Kong which are captured from 2012 to 2016. The results of the study demonstrate that neural networks approach for predicting deforestation can be utilized and its outcomes show the areas that destroyed during the research period.
ARTICLE | doi:10.20944/preprints201812.0218.v1
Subject: Physical Sciences, Mathematical Physics Keywords: approximation; ground state; neural network quantum state
Online: 18 December 2018 (10:39:29 CET)
The many-body problem in quantum physics originates from the difficulty of describing the non-trivial correlations encoded in the exponential complexity of the many-body wave function. Motivated by the Giuseppe Carleo's work titled solving the quantum many-body problem with artificial neural networks [Science, 2017, 355: 602], we focus on finding the NNQS approximation of the unknown ground state of a given Hamiltonian $H$ in terms of the best relative error and explore the influences of sum, tensor product, local unitary of Hamiltonians on the best relative error. Besides, we illustrate our method with some examples.
ARTICLE | doi:10.20944/preprints201811.0579.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Deep learning, Cognitive, LSTM, Neural network, Ngrams
Online: 26 November 2018 (10:06:05 CET)
Cognitive neuroscience is the study of how the human brain functions on tasks like decision making, language, perception and reasoning. Deep learning is a class of machine learning algorithms that use neural networks. They are designed to model the responses of neurons in the human brain. Learning can be supervised or unsupervised. Ngram token models are used extensively in language prediction. Ngrams are probabilistic models that are used in predicting the next word or token. They are a statistical model of word sequences or tokens and are called Language Models or Lms. Ngrams are essential in creating language prediction models. We are exploring a broader sandbox ecosystems enabling for AI. Specifically, around Deep learning applications on unstructured content form on the web.
ARTICLE | doi:10.20944/preprints201805.0015.v1
Subject: Business, Economics And Management, Finance Keywords: deep neural nets; market efficiency; market prediction
Online: 2 May 2018 (08:12:01 CEST)
We examine the use of deep learning (neural networks) to predict the movement of the S&P 500 Index using past returns of all the stocks in the index. Our analysis finds that the future direction of the S&P 500 index can be weakly predicted by the prior movements of the underlying stocks in the index. Decomposition of the prediction error indicates that most of the lack of predictability comes from randomness and only a little from nonstationarity. We believe this is the first test of S&P500 market efficiency that uses a very large information set, and it extends the domain of weak-form market efficiency tests.
ARTICLE | doi:10.20944/preprints201803.0118.v1
Subject: Chemistry And Materials Science, Theoretical Chemistry Keywords: machine learning; biochemistry; QSAR; molecules; neural networks
Online: 15 March 2018 (06:37:58 CET)
Biofilms are congregations of bacteria on a surface, and they grow into obstacles for the functionalities of any device or machinery involves anything biological. Biofilms are developed through a biochemical system known as ‘Quorum Sensing’ that accounts for the chemical signaling that direct either biofilm formation or inhibition. Computational models that relate chemical and structural features of compounds to their performance properties have been used to aide in the discovery of active small molecules for many decades. These quantitative structure-activity relationship (QSAR) models are also important for predicting the activity of molecules that can have a range of effectiveness in biological systems. This study uses QSAR methodologies combined with and different machine learning algorithms to predict and assess the performance of several different compounds acting in Quorum Sensing. Through computational probing of the quorum sensing molecular interaction, new design rules can be elucidated for countering biofilms.
REVIEW | doi:10.20944/preprints201705.0052.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: MEMS; microelectrodes; neural interface; conducting polymer; nanotechnology
Online: 8 May 2017 (08:39:35 CEST)
With the rapid development of MEMS (Micro-electro-mechanical Systems) fabrication technologies, manifolds microelectrodes with various structures and functions have been designed and fabricated for applications in biomedical research, diagnosis and treatment through electrical stimulation and electrophysiological signal recording. The flexible MEMS microelectrodes exhibit multi-aspect excellent characteristics beyond stiff microelectrodes based on silicon or SU-8, which comprising: lighter weight, smaller volume, better conforming to neural tissue and lower fabrication cost. In this paper, we mainly reviewed key technologies on flexible MEMS microelectrodes for neural interface in recent years, including: design and fabrication technology, flexible MEMS microelectrodes with fluidic channels and electrode-tissue interface modification technology for performance improvement. Furthermore, the future directions of flexible MEMS microelectrodes for neural interface were described including transparent and stretchable microelectrodes integrated with multi-aspect functions and next-generation electrode-tissue interface modifications facilitated electrode efficacy and safety during implantation. Finally, the combinations among micro fabrication techniques with biomedical engineering and nanotechnology represented by flexible MEMS microelectrodes for neural interface will open a new gate to human lives and understanding of the world.
ARTICLE | doi:10.20944/preprints202305.1193.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: Acoustic Emission Monitoring; Capsule Neural Network; Dilated Convolutional Neural 20 Network; Tiny Machine Learning; Time of Arrival Estimation
Online: 17 May 2023 (05:17:33 CEST)
The timely diagnosis of defects at their incipient stage of formation is crucial to extend the life-cycle of technical appliances. This is the case of mechanical-related stress, either due to long aging degradation processes (e.g., corrosion) or in-operation forces (e.g., impact events), which might provoke detrimental damages, such as cracks, disbonding or delaminations, most commonly followed by the release of acoustic energy. The localization of these sources can be successfully fulfilled via adoption of Acoustic Emission (AE)-based inspection techniques through the computation of the Time of Arrival (ToA), namely the time at which the induced mechanical wave released at the occurrence of the acoustic event arrives to the acquisition unit. However, the accurate estimation of the ToA may be hampered by poor Signal-to-Noise ratios (SNRs). In these conditions, standard statistical methods typically fail. In this work, two alternative Deep Learning methods are proposed for ToA retrieval, namely a Dilated Convolutional Neural Network (DilCNN) and a Capsule Neural Network for ToA (CapsToA). These methods have the additional benefit of being portable on resource-constrained microprocessors. Their performance has been extensively studied on both synthetic and experimental data, focusing on the problem of ToA identification for the case of a metallic plate. Results show that the two novel methods can achieve localization errors which are up to 70% more precise than those yielded by conventional strategies, even when the SNR is severely compromised (i.e., down to 2 dB). Moreover, DilCNN and CapsNet have been implemented in a tiny machine learning environment and then deployed on microcontroller units, showing a negligible loss of performance with respect to offline realizations.
ARTICLE | doi:10.20944/preprints202210.0224.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: multilabel; ensemble; incorporating multiple clustering centers; gated recurrent neural networks; temporal convolutional neural networks; long short-term memory
Online: 17 October 2022 (04:06:31 CEST)
Multilabel learning goes beyond standard supervised learning models by associating a sample with more than one class label. Among the many techniques developed in the last decade to handle multilabel learning best approaches are those harnessing the power of ensembles and deep learners. This work proposes merging both methods by combining a set of gated recurrent units, temporal convolutional neural networks, and long short-term memory networks trained with variants of the Adam optimization approach. We examine many Adam variants, each fundamentally based on the difference between present and past gradients, with step size adjusted for each parameter. We also combine Incorporating Multiple Clustering Centers and a bootstrap-aggregated decision trees ensemble, which is shown to further boost classification performance. In addition, we provide an ablation study for assessing the performance improvement that each module of our ensemble produces. Multiple experiments on a large set of datasets representing a wide variety of multilabel tasks demonstrate the robustness of our best ensemble, which is shown to outperform the state-of-the-art. The MATLAB code for generating the best ensembles in the experimental section will be made available at https://github.com/LorisNanni.
ARTICLE | doi:10.20944/preprints202001.0345.v2
Subject: Biology And Life Sciences, Biophysics Keywords: stream of consciousness; quantum mechanics; decoherence theory; synaptic transmission; spike; action potential; neural code; neural correlate of consciousness
Online: 4 March 2020 (04:54:40 CET)
In previous work, a quantum mathematical formalism associated an element of experience with a single sensory neuron, as a local reduction of a global mental state. In contrast to the binary objective states of neuronal polarisation/depolarisation, neuronal experience was modeled as a continuous variable, the instantaneous value of which could only be estimated statistically from an ensemble of evoked responses to stereotyped stimulus presentation. In the present work, the quantum operations formalism of energy dissipation through amplitude damping is adopted to explain how smooth evolution of conscious experience might arise from discrete spikes and discontinuous synaptic transmission between neurons.
ARTICLE | doi:10.20944/preprints202207.0370.v1
Subject: Medicine And Pharmacology, Oncology And Oncogenics Keywords: breast cancer; passive microwave radiometry (MWR); cascaded correlation neural network (CCNN); weight agnostic neural network (WANN); CMA-ES algorithm.
Online: 25 July 2022 (10:04:03 CEST)
Abstract Background and Objective: Medical Microwave Radiometry (MWR) is used to capture the thermal properties of internal tissues and has usages in breast cancer detection. Our goal in this paper is to improve classification performance and investigate automated neural architecture search methods. Methods: We investigate optimizing the weights of a weight agnostic neural network using bi-population covariance matrix adaptation evolution strategy (BIPOP-CMA-ES) once the topology is found. We compare it against a weight agnostic and cascade correlation neural network. Results: The experiments are conducted on a breast cancer dataset of 4912 patients. Our proposed weight agnostic BIPOP-CMA-ES model achieved the best performance. It obtained an F1-score of 0.9225, accuracy of 0.9219, precision of 0.9228, recall of 0.9217 and topology of 153 connections. Conclusions: The results are an indication of the potential of MWR utilizing a neural network-based diagnostic tool for cancer detection. By separating the tasks of topology search and weight training, we are able to improve the overall performance.
ARTICLE | doi:10.20944/preprints202306.0552.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Convolutional Neural Network; dolphin whistle; ensemble; spectrogram classification
Online: 7 June 2023 (12:54:44 CEST)
To effectively preserve marine environments and manage endangered species, it is necessary to employ efficient, precise, and scalable solutions for environmental monitoring. Ecoacoustics provides several benefits as it enables non-intrusive, prolonged sampling of environmental sounds, making it a promising tool for conducting biodiversity surveys. However, analyzing and interpreting acoustic data can be time-consuming and often demands substantial human supervision. This challenge can be addressed by harnessing contemporary methods for automated audio signal analysis, which have exhibited remarkable performance due to advancements in deep learning research. This paper introduces a research investigation into developing an automatic computerized system to detect dolphin whistles. The proposed method utilizes a fusion of various resnet50 networks integrated with data augmentation techniques. Through extensive experiments conducted on a publically available benchmark, our findings demonstrate that our ensemble yields significant performance enhancements across all evaluated metrics. The MATLAB/PyTorch source code is freely available at: https://github.com/LorisNanni/
ARTICLE | doi:10.20944/preprints202306.0260.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: chromosome classification; convolutional neural networks; ensemble; data augmentation
Online: 5 June 2023 (08:02:04 CEST)
Object classification is a crucial task in deep learning, which involves the identification and categorization of objects in images or videos. Although humans can easily recognize common objects, such as cars, animals, or plants, performing this task on a large scale can be time-consuming and error-prone. Therefore, automating this process using neural networks can save time and effort while achieving higher accuracy. Our study focuses on the classification step of human chromosome karyotyping, an important medical procedure that helps diagnose genetic disorders. Traditionally, this task is performed manually by expert cytologists, which is a time-consuming process that requires specialized medical skills. Therefore, automating it through deep learning can be immensely useful. To accomplish this, we implemented and adapted existing preprocessing and data augmentation techniques to prepare the chromosome images for classification. We used ResNet-50 convolutional neural networks and an ensemble approach to classify the chromosomes, obtaining state-of-the-art performance in the tested dataset.
ARTICLE | doi:10.20944/preprints202305.1018.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Activation Function; Neural Network; Deep Learning; RELU; SWISH
Online: 15 May 2023 (10:13:00 CEST)
Selecting the optimal activation function for training deep neural networks has always been challenging because it significantly impacts the neural network’s performance and training speed. At this point, researchers are more likely to employ RELU than other well-known activation functions. After RELU, researchers have proposed many activation functions to improve RELU. None of them was capable of surpassing RELU as their most significant rival. SWISH outperformed RELU in several challenging experiments like classification, object detection, and tracking. Replacing RELU units with SWISH, which improves performance in many tasks. This paper proposed a new activation function surpassing Google’s brain’s SWISH function, Which we named AIF. Experiments indicate that our proposed activation function outperforms SWISH in various tasks.
ARTICLE | doi:10.20944/preprints202303.0460.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: ensemble learning; bimodal; sentiment analysis; neural network; transformer
Online: 27 March 2023 (10:51:03 CEST)
Communication is a key method of expressing one's thoughts and opinions. Amongst many modalities, speech and writing are the most powerful and common forms of human communication. Analysing what and how people think has inherently been an interesting and progressive research domain. This includes bimodal sentiment analysis which is an emerging area in natural language processing (NLP) and has received a great deal of attention in recent years in a variety of areas including social opinion mining, health care, banking, and so on. At present, there are limited studies on bimodal conversational sentiment analysis as it proves to be a challenging area given the complex nature of the way humans express sentiment cues across various modalities. To address this gap, a comparison of the performance of multiple data modality models has been conducted on the MELD dataset, a widely-used dataset for benchmarking sentiment analysis within the research community. Our work then demonstrates the results of combining acoustic and linguistic representations. Lastly, our proposed neural network-based ensemble learning technique is employed over six transformer and deep learning-based models, achieving a State-Of-The-Art (SOTA) accuracy.
HYPOTHESIS | doi:10.20944/preprints202204.0025.v2
Subject: Physical Sciences, Thermodynamics Keywords: Neural synchrony; Cognition; Consciousness; Entropy; Equilibrum; Energy gradients
Online: 6 March 2023 (14:21:13 CET)
Our purpose is to address the biological problem of finding foundations of the organization in the collective activity among cell networks in the nervous system, at the meso/macroscale, giving rise to cognition and consciousness. But in doing so, we encounter another problem related to the interpretation of methods to assess the neural interactions and organization of the neurodynamics, because thermodynamic notions, which have precise meaning only under specific conditions, have been widely employed in these studies. The consequence is that apparently contradictory results appear in the literature, but these contradictions diminish upon the considerations of the specific circumstances of each experiment. After clarifying some of these controversial points and surveying some experimental results, we propose that a necessary condition for cognition/consciousness to emerge is to have available enough energy, or cellular activity; and a sufficient condition is the multiplicity of configurations in which cell networks can communicate, resulting in non-uniform energy distribution, the generation and dissipation of energy gradients due to the constant activity. These ideas may reveal possible fundamental principles of brain organization and how healthy activity may derive to pathological states.
ARTICLE | doi:10.20944/preprints202302.0026.v1
Subject: Computer Science And Mathematics, Computer Vision And Graphics Keywords: Graph Neural Network; Variational Autoencoder; Pooling; Nearest Neighbours
Online: 2 February 2023 (03:36:17 CET)
We present a Deep Learning generative model specialized to work with graphs with a regular geometry. It is build on a Variational Autoencoder framework and employs Graph convolutional layers in both encoding and decoding phases. We also introduce a pooling technique (ReNN-Pool), used in the encoder, that allows to downsample graph nodes in a spatially uniform and highly interpretable way. In the decoder, a symmetrical un-pooling technique is used to retrieve the original dimensionality of graphs. Performance of the model are tested on the standard Sprite benchmark dataset, a set of 2D images of video game characters, adequately transforming images data into graphs, and on the more realistic use-case of a dataset of cylindrical-shaped graph data that describe the distributions of the energy deposited by a particle beam in a medium.
ARTICLE | doi:10.20944/preprints202301.0406.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: AI; Sustainability; Energy efficiency; Deep learning; Neural networks
Online: 23 January 2023 (09:38:57 CET)
As AI models become more and more common in business and even in our daily lives, it is important to understand what the carbon impact of these models is. Recent papers have shown that this impact can be quite great, i.e., the training of a single high-end model can result in emissions of more than 500t of CO2eq. In this paper we discuss the factors that influence the carbon footprint of AI models, explore what impact different decisions have, and show how the footprint can be reduced. We also examine the footprint of different models to give a guideline on how urgent action is for different organizations.
ARTICLE | doi:10.20944/preprints202211.0557.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: hydroelectric basin modelling; spatial interpolation; neural networks; Kriging
Online: 30 November 2022 (03:10:29 CET)
In this work a new hydroelectric basin modelling approach is described and applied to the Pontecosi basin, Italy. Several types of data sources were used to learn the model: a number of weather stations, satellite observations, Reanalysis dataset and basin data. With the goal of predicting the water level of the basin, the model was composed by three cascade modules. Firstly, different spatial interpolation methods, such as Kriging, Radial Basis Function and Natural Neighbours, were compared and applied to interpolate the weather stations data nearby the basin area to infer the main environmental variables (air temperature, air humidity, precipitation and wind speed) in the basin area. Then, using these variables as inputs, a neural network was trained to predict the mean soil moisture concentration over the area, also to improve the low availability due to satellite orbits. Finally, a non-linear auto regressive exogenous input (NARX) model was trained to simulate the basin level with different prediction horizons, using the data from the previous modules and past basin data (water level, discharge flow rate, turbine flow rate). Accurate predictions of the basin water level were achieved within 1 to 6 hours ahead, with mean absolute errors (MAE) between 2cm and 10cm respectively.
REVIEW | doi:10.20944/preprints202210.0391.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Tillage; Traction; Compaction; Neural networks; Support vector regression
Online: 26 October 2022 (02:07:19 CEST)
Soil working tools, implements, and machines are inevitable in mechanized agriculture. The soil-tool/machine interaction is a multivariate, dynamic, and intricate process. The accurate interpretation, description, and modeling of a soil-machine interaction is key to providing a solution to sustainable crop production by reducing energy input, excessive soil pulverization, and compaction. The traditional method provides insight into soil-machine interaction but often provides inadequate solutions and lacks broad applicability. Computational intelligence (CI) is a comprehensive class of approaches that rely on approximate information to solve complex problems. The CI method has been extensively studied and applied in soil tillage and traction domain in recent decades. The study critically reviews the CI techniques implemented in soil-machine interactions, especially in the context of tillage, traction, and compaction. The traditional methods and their limitation are discussed. The fundamental of CI methods and a detailed overview of the most popular methods are provided. The study reviews and summarizes the 50 selected articles on soil-machine interaction studies where CI methods were employed. It discusses the strength and limitations of employed CI methods. It also suggests the emergent CI methods and future applications are discussed. The outlined study would serve as a concise reference and a quick and systematic way to understand the applicable CI methods that allow crucial farm management decision-making.
ARTICLE | doi:10.20944/preprints202209.0415.v1
Subject: Computer Science And Mathematics, Computational Mathematics Keywords: Stylometry; Signal Processing; Word Embedding; Deep Neural Networks
Online: 27 September 2022 (07:55:25 CEST)
Classical religious texts remain an essential part of human culture due to their undiminished influence on the advancement of civilization. Although their entirely divine origin is questioned repeatedly, explicit or implicit quoting and adherence to their basic guidelines are fundamental in modern society. In this respect, these documents’ inner structure and linguistic style appear to be pivotal. This paper considers the topic from the standpoint of small textual patterns classified using deep learning methods, traditionally applied to analyze short textual material like tweets. We divide the considered documents into small sequential chunks imitating tweets and categorizing them, classifying an entire text. The proposed method demonstrates that the religious text collections correspond to stable ”Twitter”-like structures that adequately reflect stylistic properties. So, concise word combinations seem to be an inborn textual attribute that adequately outlines the proposed multi-source authorship. This approach differs from traditional methods of analyzing classical religious documents, which are based on the consideration and interpretation of relatively long templates. The case study consists of three famous collections of Mosaic authorship in the Old Testament (Hebrew), Pauline authorship in the New Testament (Greek), and Al-Ghazali authorship (Arabic). The obtained results go well with most previously expressed evaluations and complement them with new implications, particularly in the authorship of two famous manuscripts attributed to Al-Ghazali.
COMMUNICATION | doi:10.20944/preprints202209.0041.v1
Subject: Computer Science And Mathematics, Artificial Intelligence And Machine Learning Keywords: Deep Learning; Convolutional Neural Networks; Medical Image Segmentation
Online: 5 September 2022 (03:12:55 CEST)
Convolutional neural network architectures have become increasingly complex, which has improved the performance slowly on well-known benchmark datasets in the recent years. In this research, we have analyzed the true need for such complexity. We have introduced G-Net light, a lightweight modified GoogleNet with improved filter count per layer to reduce feature overlaps and complexity. Additionally, by limiting the amount of pooling layers in the proposed architecture, we have exploited the skip connections to minimize the spatial information loss. The investigations on the proposed architecture are evaluated on three retinal vessel segmentation publicly available datasets. The proposed G-Net light outperforms other vessel segmentation architectures by reducing the number of trainable parameters..