Wernicke-Korsakoff syndrome (WKS) is induced by thiamine deficiency (TD) and mainly related to alcohol consumption. Frontal cortex dysfunction has been associated to impulsivity and disinhibition in WKS patients. The pathophysiology involves oxidative stress, excitotoxicity and inflammatory responses leading to neuronal death, but the relative contributions of each factor (alcohol and TD, isolate or in interaction) to these phenomena are still poorly understood. A rat model was used by forced consumption of 20% (w/v) alcohol for 9 months (CA), TD hit (TD diet + pyrithiamine 0.25 mg/kg, i.p. daily injections the last 12 days of experimentation; TDD), and both combined treatments (CA+TDD). Motor and cognitive performance and cortical damage were examined. CA caused hyperlocomotion as a possible sensitization of ethanol-induced excitatory effects and recognition memory deficits. In addition, CA+TDD animals showed a disinhibited-like behavior, which appears to be dependent on TDD. Also, combined treatment led to more pronounced alterations in nitrosative stress, lipid peroxidation, apoptosis and cell damage markers. Correlations between injury signals and disinhibition suggest that CA+TDD disrupts behaviors dependent on the frontal cortex. Our study sheds light on the potential disease-specific mechanisms, reinforcing the need for neuroprotective therapeutic approaches along with preventive treatments for the nutritional deficiency in WKS.

Publicly-funded scientists have a responsibility to engage with the public on scientific information, but are lacking a standardized framework and assessment strategy to do it well. The PEPS (Public Engagement Practices for Scientists) Method is an outcomes-centered framework employing standardized pedagogical methods with quantifiable outcomes. This approach reveals that scientists often have unrealistic expectations for achieving affective learning outcomes (i.e. changing views from anti- to pro-vaccine) by solely cognitive learning strategies (i.e. supplying data). The PEPS Method can serve as a roadmap for standardized science communication assessments, complementing existing science communication training programs for the next generation of scientists.

Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research towards the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we introduce a set of metrics considering both quantitative and qualitative aspects of the road network. Compared to traditional neural networks, our HAN model demonstrates superior accuracy and F1-score, particularly for road types with comparable samples. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.

Land cover segmentation, a fundamental task within the domain of remote sensing, boasts a broad spectrum of application potential. In this article, we focus on the land cover segmentation tasks and complete the following research work: Firstly, to address the issues of uneven distribution of foreground and background and significant differences in target scales in remote sensing images, we propose a decoder called MDCFD based on multi-dilation rate convolution fusion. The decoder utilizes dilated convolution to expand the receptive field, enhancing the model's ability to capture global features and thus improving the model's ability to distinguish between foreground and background. Meanwhile, we design a multi-dilation rate convolution fusion module (MDCFM), which fuses the outputs of different dilation rate convolution layers. Secondly, aiming at the problems of diverse scenes, significant differences between categories, and many background interferences in remote sensing images, we propose a hybrid attention module called LKSHAM based on large kernel convolution. This module combines spatial attention and channel attention mechanisms and combines the two attention modules in series. By introducing large kernel convolution, the model's ability to extract contextual information is improved. At the same time, we adopt a strategy of decomposing large kernel convolution into multiple depthwise convolutions to reduce computational complexity. The improved network models designed by this paper can achieve an improvement of over 1.1% in the mIoU metric on segmentation tasks on the Potsdam and Vaihingen datasets.

Parkinson’s Disease (PD) is a neurodegenerative disorder characterized by motor and non-motor impairment with various implications on patients’ quality of life. Since currently available therapies are only symptomatic, identifying individuals with prodromal, preclinical or early-stage PD is crucial, as they would be ideal candidates for future disease-modifying therapies. Our analysis aims to develop a robust model for accurate PD detection using accelerometer data collected from PD and non-PD individuals with mild or no tremor during phone conversations. An open-access dataset comprising accelerometer recordings from 22 PD patients and 11 healthy controls (HC) was utilized. The data were pre-processed to extract relevant time, frequency and energy-related features, and a Bidirectional Long Short-Term Memory (Bi-LSTM) model with attention mechanism was employed for classification. The performance of the model was evaluated using 5-fold cross-validation, and metrics of accuracy, precision, recall, specificity, and f1-score were computed. The proposed model demonstrated high accuracy (98%), precision (99%), recall (98%), specificity (96%), and f1-score (98%) in accurately distinguishing PD patients from HC. Our findings indicate that the proposed model outperforms existing approaches and holds promise for detection of PD with subtle symptoms, like tremor, in the wild. Such symptoms can present in the early or even prodromal stage of the disease, and appropriate mobile-based applications may be a practical tool in real-life settings to alert individuals at risk to seek medical assistance or give patients feedback in monitoring their symptoms.

Dialogue State Tracking is an essential component in multi-domain dialogue systems, aiming to accurately determine the current dialogue state based on the dialogue history. Existing research has addressed the issue of multiple mappings in dialogues by employing slot self-attention as a data-driven approach. However, learning the relationships between slots from a single sample has limitations and introduces noise. In this paper, we propose an External Slot Relation Memory-based Dialogue State Tracking model (ER-DST). By utilizing an external memory storage, we learn the relationships between slots as a dictionary of multi-domain slot relations. Additionally, we employ a small filter to discard slot information irrelevant to the current dialogue state. Experimental results on the MultiWOZ2.0 and MultiWOZ2.1 benchmarks demonstrate significant improvements while reducing the time complexity to O(n).

Human pose estimation is a complex detection task in which the network needs to capture the rich information contained in the images. In this paper, we propose MSTPose (Multi-Scale Transformer for human Pose estimation). Specifically, MSTPose leverages a high-resolution Convolution neural network (CNN) to extract texture information from images. For the feature maps from three different scales produced by the backbone network, each branch performs the coordinate attention operations. The feature maps are then spatially and channel-wise flattened, combined with keypoint tokens generated through random initialization, and fed into a parallel Transformer structure to learn spatial dependencies between features. As the Transformer outputs one-dimensional sequential features, the mainstream two-dimensional heatmap method is abandoned in favor of one-dimensional coordinate vector regression. The experiments show that MSTPose outperforms other CNN-based pose estimation models and demonstrates clear advantages over CNN + Transformer networks of similar types.

Teeth segmentation plays a pivotal role in dental diagnosis, treatment, planning, and the development of computer-aided dental systems. It enables precise identification and analysis of dental structures, aiding in detecting dental abnormalities, measuring tooth dimensions, and assessing oral health conditions. Accurate teeth segmentation also facilitates the automation of dental workflows, leading to improved efficiency and reduced human error. Artificial Intelligence (AI) has witnessed rapid advancements, with various approaches developed or enhanced. While Convolutional Neural Networks (CNNs) have been widely used in medical image segmentation, the U-Net architecture has emerged as a standout performer due to its exceptional segmentation capabilities. This paper presents a proof of concept for the Attention U-Net architecture, as previously used in [1], applied to teeth segmentation. The study demonstrates the superior performance of this network in accurately segmenting teeth using a newly available benchmark dataset called Tufts Dental X-Ray Dataset. When trained and tested on 10-fold cross-validation, the model achieved an average dice coefficient of 95.01%, intersection over union of 90.6%, and pixel accuracy of 98.82%. These scores surpass those of all other networks implemented on the same dataset. By leveraging the Attention U-Net architecture, our research showcases the potential of advanced AI techniques in dental radiography. The findings contribute to the ongoing efforts to develop automated systems to assist dental professionals in their clinical practice.

With the rapid development of Internet of Things (IoT)-based near-earth remote sensing technology, the problem of network intrusion for near-earth remote sensing systems has become more complex and large-scale. Therefore, it is essential to seek an intelligent, automated, and robust network intrusion detection method. In recent years, network intrusion detection methods based on graph neural networks (GNNs) have been proposed. However, there are still some practical issues with these methods. For example, they have not taken into consideration the characteristics of near-earth remote sensing systems, the state of the nodes, and the temporal features. Therefore, this article analyzes the characteristics of existing near-earth remote sensing systems and proposes a spatio-temporal graph attention network (N-STGAT) that considers the state of nodes. The proposed network applies spatiotemporal graph neural networks to the network intrusion detection of near-earth remote sensing systems and validates the effectiveness of the proposed method on the latest flow-based dataset.

Culture exploits the acquisition of meaningful content by crafting regimes of shared attention, determining what is relevant, valuable, and salient. Culture changes the field of relevant social affordances worthy of being acted upon in a context-sensitive manner. When relevant affordances are highly weighted, their attentional capture and their salience increase the probability of them being enacted due to the associated expectation for minimizing prediction error. This process is known as active inference. In the digital era, individuals need to infer the action-related attributes of digital cues, here characterized as digital affordances. The digital affordances of digital social platforms are of particular interest here. Digital social affordances are defined as online possibilities of social interactions. By their own nature, these are salient because they are related to social interactions and relevant social cues. However, the problem of digital social platforms is that they are not equivalent to situated social interactions because their structure is built, mediated, and defined by third-parties with diverse interests. The third-parties behind the digital social platforms are using the same mechanism exploited by culture to manipulate the shared patterns of attention. Moreover, digital social platforms are deliberately designed to be hyper-stimulating, making digital social affordances highly rewarding and increasingly salient. This appropriation, for economic purposes, is an issue of great importance, especially as the COVID-19 pandemic brought deep global changes, pushing societies to an online digital way of life. Here, we examined different types of digital social affordances under an active inference view, placing them into two categories, those for self-identity formation, and those for belief-updating. This paper aims to analyze digital social affordances in light of the prediction error dynamics they might elicit to their users. Although each of the analyzed digital social affordances allows different epistemic and instrumental digital actions, they all share the characteristic of having an "easy" and a fast expected rate of error reduction. Here, we aim to provide a new hypothesis about how the design behind digital social affordances is built on our natural attractiveness to minimize prediction error and the resulting positive embodied feelings when doing so. Finally, it is suggested that because digital social affordances are becoming highly weighted in the field of affordances, this might be putting at risk our context-sensitive grip on a rich, dynamic and varied field of relevant affordances.

The object detection task is usually affected by complex backgrounds. In this paper, a new image object detection method is proposed, which can perform multi-feature selection on multi-scale feature maps. By this method, a bidirectional multi-scale feature fusion network is designed to fuse semantic features and shallow features to improve the detection effect of small objects in complex backgrounds. When the shallow features are transferred to the top layer, a bottom-up path is added to reduce the number of network layers experienced by the feature fusion network, reducing the loss of shallow features. In addition, a multi-feature selection module based on the attention mechanism is used to minimize the interference of useless information on subsequent classification and regression, allowing the network to adaptively focus on appropriate information for classification or regression to improve detection accuracy. Because the traditional five-parameter regression method has severe boundary problems when predicting objects with large aspect ratios, the proposed network treats angle prediction as a classification task. The experimental results on the DOTA dataset, the self-made DOTA-GF dataset and the HRSC 2016 dataset show that, compared with several popular object detection algorithms, the proposed method has certain advantages in detection accuracy.

Satellite communication is expected to play a vital role in realizing Internet of Remote Things (IoRT) applications. This article considers an intelligent reflecting surface (IRS)-assisted downlink low Earth orbit (LEO) satellite communication network, where IRS provides additional reflective links to enhance the intended signal power. We aim to maximize the sum-rate of all the terrestrial users by jointly optimizing the satellite’s precoding matrix and IRS’s phase shifts. However, it is difficult to directly acquire the instantaneous channel state information (CSI) and optimal phase shifts of IRS due to the high mobility of LEO and the passive nature of reflective elements. Moreover, most conventional solution algorithms suffer from high computational complexity and are not applicable to these dynamic scenarios. A robust beamforming design based on graph attention networks (RBF-GAT) is proposed to establish a direct mapping from the received pilots and dynamic network topology to the satellite and IRS’s beamforming, which is trained offline using the unsupervised learning approach. The simulation results corroborate that the proposed RBF-GAT can achieve approximate performance compared to the upper bound with low complexity.

For any molecule, network, or process of interest, to keep up with new publications on these, is becoming increasingly difficult. For many cellular processes, molecules and their interactions that need to be considered can be very large. Automated mining of publications can support large scale molecular interaction maps and database curation. Text mining and Natural Language Processing (NLP)-based techniques are finding their applications in mining the biological literature, handling problems such as Named Entity Recognition (NER) and Relationship Extraction (RE). Both rule-based and machine learning (ML)-based NLP approaches have been popular in this context, with multiple research and review articles examining the scope of such models in Biological Literature Mining (BLM). In this review article, we explore self-attention based models, a special type of neural network (NN)-based architectures that have recently revitalized the field of NLP, applied to biological texts. We cover self-attention models operating either at a sentence level or an abstract level, in the context of molecular interaction extraction, published from 2019 onwards. We conduct a comparative study of the models in terms of their architecture. Moreover, we also discuss some limitations in the field of BLM that identifies opportunities for the extraction of molecular interactions from biological text.

Alcohol use disorder (AUD) is a chronic and relapsing biopsychosocial condition that leads to a wide range of medical, psychological, social, economic, and personal issues. The study investigated the predictors of quality of life (QoL) among male patients seeking treatment for AUD based on attention, cognitive instability, motor impulsiveness, perseverance, and self-control dimensions of trait impulsivity. The study also investigated the potential mediating effect of anxiety and alcohol use severity in this relationship. After an initial assessment, participants underwent comprehensive evaluations, including sociodemographic data, clinical history, and QoL domains. Evaluations also covered AUD severity, impulsivity dimensions, cognitive functions, craving intensity, depression, anxiety and self-efficacy. Multiple linear regressions were employed to identify associations between independent variables and QoL. Attention dimension of impulsivity and anxiety symptoms significantly predicted physical health QoL. Self-control dimension and severity of alcohol use disorder predicted psychological QoL. Indirect effect of attention dimension on physical health was significant [effect =-1.082,95% CI (-2.008, -0.3598)].The significance of impulsivity in influencing QoL for individuals with AUD has profound clinical implications. These findings underscore the importance of addressing anxiety and impulsivity in managing AUD, given their considerable impact, particularly on QoL, ultimately shaping treatment outcomes.

The noradrenergic theory of Cognitive Reserve (Robertson, 2013-2014) postulates that the upregulation of the Locus Coeruleus - Noradrenergic System (LC-NA) originating in the Brainstem might facilitate cortical networks involved in attention, and protracted activation of this system throughout the lifespan may enhance cognitive stimulation contributing to Reserve. To test the above-mentioned theory, a study was conducted on a sample of 686 participants (395 controls, 156 Mild Cognitive Impairment, 135 Alzheimer’s Disease) investigating the relationship between LC volume, attentional performance and a biological index of brain maintenance (BrainPAD – an objective measure which compares an individual’s structural brain health, reflected by their voxel-wise grey matter density, to the state typically expected at that individual’s age). Further analyses were carried out on Reserve indices including education and occupational attainment. Volumetric variation across groups was also explored along with gender differences. Control analyses on the Serotoninergic (5-HT), Dopaminergic (DA) and Cholinergic (Ach) systems were contrasted with the Noradrenergic (NA) hypothesis. The antithetic relationships were also tested across the neuromodulatory subcortical systems.Results supported by bayesian modelling showed that LC volume disproportionately predicted higher attentional performance as well as biological brain maintenance across the three groups. These findings lend support to the role of the noradrenergic system as a key mediator underpinning the neuropsychology of Reserve, and they suggest that early prevention strategies focused on the noradrenergic system (e.g. cognitive-attentive training, physical exercise, pharmacological and dietary interventions) may yield important clinical benefits to mitigate cognitive impairment with age and disease.

In the remote sensing domain, it is crucial to complete semantic segmentation on the raster images, e.g., river, building, forest, etc, on raster images. A deep convolutional encoder--decoder (DCED) network is the state-of-the-art semantic segmentation method for remotely sensed images. However, the accuracy is still limited, since the network is not designed for remotely sensed images and the training data in this domain is deficient. In this paper, we aim to propose a novel CNN for semantic segmentation particularly for remote sensing corpora with three main contributions. First, we propose applying a recent CNN called a global convolutional network (GCN), since it can capture different resolutions by extracting multi-scale features from different stages of the network. Additionally, we further enhance the network by improving its backbone using larger numbers of layers, which is suitable for medium resolution remotely sensed images. Second, "channel attention'' is presented in our network in order to select the most discriminative filters (features). Third, "domain-specific transfer learning'' is introduced to alleviate the scarcity issue by utilizing other remotely sensed corpora with different resolutions as pre-trained data. The experiment was then conducted on two given datasets: (i) medium resolution data collected from Landsat-8 satellite and (ii) very high resolution data called the ISPRS Vaihingen Challenge Dataset. The results show that our networks outperformed DCED in terms of $F1$ for 17.48% and 2.49% on medium and very high resolution corpora, respectively.

Remote sensing technology is crucial for accurate rooftop detection, benefiting urban planning, disaster management, and solar resource estimation. This study employs Efficient Interactive Segmentation (EISEG) to enhance the efficiency of remote sensing image labeling, with a particular focus on rooftop detection. It is necessary to use modern technology because traditional manual labelling methods are labor-intensive and complicated. The study introduces a novel framework on deep learning semantic segmentation models, facilitating an efficient approach to rooftop identification using high-resolution UAV remote sensing datasets. Large dataset of labeled UAV rooftop building images, in which each superpixel region is assigned a binary label indicating rooftop presence. Advanced methods including Asymmetric Neural Network (ANN), Dual At-tention Network (DANet), PP-LiteSeg, and Deeplab3 are implemented for automatic rooftop detection due to their higher performance and advanced architectures. These models are executed on the Baidu deep learning platform PaddlePaddle, generating initial rooftop segmentation maps crucial for estimating photovoltaic resources. The ANN model emerges with the highest accuracy at 96%, followed by DANet at 95.09%, PP-LiteSeg at 94.54%, and Deeplab3 at 81.61%. The outcomes presenting efficient models for automated rooftop identification, and demonstrating the continuous need for improving deep learning techniques in smart and sustainable cities.

Real-time object detection remains a pivotal topic in the realm of computer vision. Balancing the accuracy and speed of object detectors poses a formidable challenge for both academic researchers and industry practitioners. While recent transformer-based models have showncase the prowess of the attention mechanism, delivering remarkable performance improvements over CNNs, their computational demands can hinder their effectiveness in real-time detection tasks. In this paper, we elect to use YOLOX as our robust starting point and introduce a series of effective enhancements, culminating in a new high-performance detector named YOLOAX. To further exploit the power of the attention mechanism, we devise multi-dimensional attention-based modules that activate CNNs, emphasizing the most pertinent regions and bolstering the capacity to learn the most informative image representations from feature maps. Moreover, we introduce a novel leading label assignment strategy called STA, along with a groundbreaking loss function named GEIOU Loss, to further refine our detector's performance. We provide extensive ablation studies on the COCO and PASCAL VOC 2012 detection datasets to validate our proposed methods. Remarkably, our YOLOAX is trained solely on the MS-COCO dataset from scratch, without leveraging any prior knowledge, which achieves an impressive 54.2% AP on the COCO 2017 test set while maintaining a real-time speed of 72.3 fps, surpassing YOLOX by a significant margin of 3.0% AP. Our source code and pre-trained models are openly accessible at https://github.com/KejianXu/yoloax.

In order to improve real-time performance of gesture recognition by micro-Doppler map of mmWave Radar, the point cloud based gesture recognition for mmWave Radar is proposed in this paper.Two steps are carried out for mmWave Radar based gesture recognition. The first step is to estimate the point cloud of the gestures by 3D-FFT and the peak grouping. The second step is to train the TRANS-CNN model by combining the multi-head self-attention and the 1D-convolutional network, so as to extract the features in the point cloud data at a deeper level to categorize the gestures. In the experiments, TI mmWave Radar sensor IWR1642 is used as benchmark to evaluate the feasibility of the proposed approach. The results show that the accuracy of the gesture recognition reaches 98.5%. In order to prove the effectiveness of our approach, a simply 2Tx2Rx Radar sensor is developed in our lab and the accuracy of recognition reaches 97.1%. The results show that our proposed gesture recognition approach achieves the best performance in real-time with limited training data, in comparison with the existing methods.

The semantic segmentation of high-resolution remote sensing images (HR-RSI) is crucial for a wide range of applications, such as precision agriculture, urban planning, natural resource assessment, and ecological monitoring. However, accurately classifying pixels in HR-RSI faces challenges due to densely distributed small objects and scale variations. Existing techniques, including Convolutional Neural Networks (CNNs) and other methods for hierarchical feature extraction and fusion of remote sensing image, often do not achieve the desired accuracy. In this paper, we propose a novel approach called the Hierarchical Rich-scale Fusion Network (HRFNet) to address these challenges. HRFNet utilizes advanced information rating and image partition techniques to extract rich-scale features within image layers. This allows for the adaptive exploration of both local and global contextual information. Moreover, we introduce a structured intra-layer to inter-layer feature aggregation module, which enables the adaptive extraction of fine-grained details and high-level semantic information from multi-layer feature maps in a highly flexible manner. Extensive experimentation has been conducted to validate the effectiveness of our proposed method. Our results demonstrate that HRFNet outperforms existing techniques, achieving state-of-the-art (SOTA) results on benchmark datasets, specifically the ISPRS Potsdam and Vaihingen datasets.

To address the problems of poor recognition, low detection accuracy, a large number of model parameters and computation, complex network structure, and unfavorable portability to embedded devices in traditional tennis ball detection algorithms, this study proposes a lightweight tennis ball detection algorithm YOLOv5s-Z based on Robomater EP. The main work is as follows: Firstly, constructing lightweight G-Backbone and G-Neck network layers to reduce the number of parameters and computation of the network structure. Secondly, the convolutional coordinate attention is incorporated in G-Backbnone to embed the location information into the channel attention, which makes the network obtain the location information in a larger area through multiple convolutions and enhances the expression ability of the mobile network learning features. In addition, the Concat module in the original feature fusion is modified into a weighted bi-directional feature pyramid W-BiFPN with settable learning weights to improve the feature fusion capability and achieve efficient weighted feature fusion and bi-directional cross-scale connectivity. The EIOU loss is introduced to split the influence factor of aspect ratio and calculate the length and width of the target frame and anchor frame respectively, combined with Focal-EIOU Loss to solve the problem of imbalance between difficult and easy samples. The activation function Meta-ACON is introduced to achieve an adaptive selection of whether to activate the neurons and improve the detection accuracy. Finally, the experimental results show that compared with the original algorithm, the YOLOv5s-Z algorithm reduces the number of parameters and computation by 42$\%$ and 44$\%$, the model size by 39$\%$, and 2$\%$ improvement in average accuracy mean value, which verifies the effectiveness of the improved algorithm and the light weight of the model to meet the deployment requirements of embedded devices, and adapts Robomaster EP for accurate detection and real-time recognition of tennis balls.

Deep convolutional neural networks (CNNs) have been widely used in various medical imaging tasks. However, due to the intrinsic locality of convolution operation, CNNs generally cannot model long-range dependencies well, which are important for accurately identifying breast cancer from unregistered multi-view mammograms. This motivates us to leverage the architecture of Vision Transformers to capture long-range relationships of multiple mammograms from the same patient. For this purpose, we employ local Transformer blocks to separately learn patch relationships within a specific view (CC/MLO) of one side (right/left) of mammogram. The outputs from different views and sides are concatenated and fed into global Transformer blocks, to jointly learn patch relationships between two different views of the left and right breasts. To evaluate the proposed model, we retrospectively assembled a dataset involving 949 sets of CC and MLO view mammograms, which include 470 malignant cases and 479 normal or benign cases. We trained and evaluated the model using a five-fold cross-validation method. Without any arduous preprocessing steps (e.g., optimal window cropping, chest wall or pectoral muscle removal, two-view image registration, etc.), our two-view Transformer-based model achieves lesion classification accuracy of 77.0% and area under ROC curve (AUC = 0.814), outperforming state-of-the-art multi-view CNNs by 3.1% and 3%, respectively. Meanwhile, the new two-view model improves mammographic case classification accuracies of two single-view models by 7.4% (CC) and 4.5% (MLO), respectively. The promising results unveil the great potential of using Transformers to develop high-performing computer-aided diagnosis (CADx) schemes of mammography.

Triple X syndrome (TXS, also known as trisomy X or 47,XXX) has been associated with impaired overall neurocognitive functioning in children and relatively young adults. However, neurocognitive functioning in adults with TXS is poorly understood. The aim of this study was, therefore, to examine cognitive functioning in adults with TXS. Methods: In this cross-sectional study, data were collected from 34 adult women with TXS (mean age = 32.9; SD = 13.1) and 31 controls (mean age = 34.9; SD = 13.7). General intellectual functioning, semantic/verbal memory, visual/episodic memory, psychomotor speed, and attention and executive functioning were then compared between these two groups. Results: We found that general intellectual functioning was significantly lower in the TXS group compared to the control group. In addition, women with TXS had more attention problems and lower psychomotor speed, particularly motor processing speed. When the analyses were adjusted for IQ, the strength of these associations decreased. The women in the TXS group also scored significantly lower at free recall in the verbal memory test, but not in immediate or delayed recognition. Finally, visual/episodic memory and executive functioning did not differ significantly between groups. Conclusions: Our analysis revealed that women with TXS score lower in general intellectual functioning and have impairments in motor processing speed and attention compared to controls, but do not differ with respect to executive functioning. These results offer new insights for improving the support of adults with TXS both at school and in the workplace.

A hemispheric asymmetry is known for the processing of global vs. local visual information. In this study, we investigated the existence of a hemispheric asymmetry for visual processing of low vs. high spatial frequency gratings. Event-related potentials were recorded in a group of healthy right-handed volunteers from 30 scalp sites. Six types of stimuli (1.5, 3 and 6 c/deg gratings) were randomly flashed 180 times in the left and right upper hemi-fields. Stimulus duration was 80 ms and ISI ranged between 850-1000 ms. Participants had to pay attention and respond to targets based on their spatial frequency and location, or to passively look at the stimuli. C1 and P1 visual responses, as well as a later Selection negativity and a P300 components of ERPs were quantified and subjected to repeated-measure ANOVAs. Overall, performance was faster for the RVF, thus suggesting a left hemispheric advantage for attentional selection of local elements. Similarly, the analysis of mean area amplitude of C1 (60-110 ms) sensory response showed a stronger attentional effect (F+L+ vs. F-L+) at left occipital areas, thus suggesting the sensory nature of this hemispheric asymmetry.

A 47-year-old physician suddenly noticed a persistent difficulty maintaining attention. He was awake, alert, and oriented. After two hours he developed fever, ageusia, and anosmia. He denied any previous history of psychiatric illness and was hydrated at the time of the subjective attention impairment. On admission, the patient remained oriented. He reported the persistence of attention problems, anosmia, and mild fatigue. The oxygen saturation 99% while he was breathing ambient air. Laboratory tests were unremarkable. A high-resolution computed tomography of the chest was normal. Nasopharyngeal and throat swabs specimens on reverse transcription-polymerase chain reaction analysis tested positive for SARS-CoV2. On illness day 3, the examination was unchanged, but he continued to complain of difficulties to stay focused. Then, he performed an objective attention test. The test demonstrated a moderate attentional impairment. On day 6, the patient reported a subjective worse in his concentration and performed a second test. Although his physical examination remained normal, the attention performance was worse as compared to day 3. Eight hours after worsening of attention impairment, the patient’s oxygen saturation dropped to 94%. From illness days 9 to 14, the patient evolved with clinical improvement. On day 10, a third objective attention test indicated a mild deficit. On day 16, he did not report any other symptom and the attention test was completely normal. Then, the patient returned to work. Neurological symptoms had been previously described in COVID- 19 patients. However, no previous research had investigated early cognitive deficits preceding the traditional symptoms.

Humans are incapable of judging the temporal order of visual events at brief temporal separations with perfect accuracy. Their performance---which is of much interest in visual cognition and attention research---can be measured with the temporal-order judgment task, which typically produces S-shaped psychometric functions. Occasionally, researchers reported plateaus within these functions, and some theories predict such deviation from the basic S shape. However, the centers of the psychometric functions result from the weakest performance at the most difficult presentations and therefore fluctuate strongly, leaving existence and exact shapes of plateaus unclear. This study set out to investigate whether plateaus disappear if the data accuracy is enhanced, or if we are ``stuck on a plateau'', or rather with it. For this purpose, highly accurate data were assessed by model-based analysis. The existence of plateaus is confidently confirmed and two plausible mechanisms derived from very different models are presented. Neither model, however, performs well in the presence of a strong attention manipulation, and model comparison remains unclear on the question which of the models describes the data best. Nevertheless, the present study includes the highest accuracy in visual TOJ data and the most explicit models of plateaus in TOJ studied so far.

As on 31.03.2020, 55.4 % (205135 MW) of total installed capacity (370106 MW) in India is through coal and lignite based power plants. These plants, set up by central, state and private utilities with substantial capital investment are facing consistently reducing Plant Utilization Factor (known as Plant Load Factor, PLF, in India). In the year 2019-20 the national average thermal power PLF stood at 55.4%, down from 78.6 % in 2007-08. On the other hand, the electricity demand is consistently rising in the country and there exists a peak and energy shortage at national level. In 2019-20 energy shortage was 0.7 % and peak shortage was 0.5 %. A disturbing paradox therefore exists here. On one hand, the country is power deficit, and on the other hand, a large amount of coal based affordable power, ready to be generated by thermal power generators, remains grossly unused. Looking into the fact that considerable investment has gone into developing these thermal power generation assets in the country, the falling PLF is a matter of concern for all the key stakeholders including the power producers, lenders, regulators and consumers. This paper identifies seven major factors that are affecting PLF of thermal power plants and then makes an attempt to project future scenario of PLF so that critical stakeholders can intervene through appropriate actions. Primary research with responses from power professionals has been used to find out the major factors. Future projection of PLF has been done using Partial Least Square (PLS) regression. Projection shows that in the Business As Usual case (Factors increasing at the current CAGR rate), the thermal power plants will face very low level of PLF (14.76 %) by 2024-25. This will mean that many plants will be shut down and many will run for only few hours in a day that too at very low loads. If the future generation mix is kept as indicated by Central Electricity Authority (CEA), a Govt. of India in its report (Draft report on optimal generation capacity mix for 2029-30- CEA- Govt of India) then the thermal power plant average PLF can sustain above 68 % until 2024-25. If followed, this path can be a breather for the thermal power plants.

Underwater robotics equipped with visual detection systems have the function of detecting un-derwater artifacts, which is of great significance to deep-sea archaeology. Underwater artifacts are located in complex environments with poor imaging conditions, and the targets have chal-lenging problems such as breakage, stacking, and sediment burial, which causes the marine cul-tural target detection failures. To solve these problems, this paper proposes an underwater cul-tural target detection algorithm based on the deformable deep aggregation network model for underwater robotics exploration. To fully extract the target feature information of underwater targets in complex environments, this paper designs a multi-scale deep aggregation network with deformable convolutional layers. Besides, the BAM attention module is designed for fea-ture optimization, which enhances the potential feature information of the target while weak-ening the background interference information. Finally, the target prediction is achieved through feature fusion at different scales. The proposed algorithm has been extensively validat-ed and analyzed on the collected underwater artifact datasets, and the precision, recall, and mAP of the algorithm have reached 93.1%, 91.4%, and 92.8%, respectively. In addition, the present al-gorithm has been practically deployed on the underwater robotics. In the deep-sea tests, the ar-tifact detection frame rate reaches up to 19 fps, which satisfies the real-time target detection。

Drugs-drugs interactions(DDI) are entities composed of different chemical substructures (functional groups). In existing methods that predict drug–drug interactions based on the usage of substructures, each node is considered the center of a substructure, and adjacent nodes eventually become centers of similar substructures, resulting in redundancy. Furthermore, the significant differ- ences in structure and properties among compounds can lead to unrelated pairings, making it difficult to integrate information. This heterogeneity negatively affects the prediction results. To address these issues, we propose a drug–drug interaction prediction method based on substructure signature learning (DDI-SSL). This method extracts useful information from local subgraphs surrounding drugs and effectively utilizes substructures to assist in predicting drug side effects. Additionally, a deep clustering algorithm is used to aggregate similar substructures, allowing any individual subgraph to be reconstructed using this set of global signatures. Furthermore, we developed a layer-independent collaborative attention mechanism to model the mutual influence between drugs, generating signal strength scores for each class of drugs to mitigate noise caused by heterogeneity. Finally, we evaluated DDI-SSL on a comprehensive dataset and demonstrated improved performance in DDI prediction compared to state-of-the-art methods.

AVs suffer reduced maneuverability and performance due to the degradation in sensor performances in fog. Such degradation causes significant object detection errors essential for AVs' safety-critical conditions. For instance, YOLOv5 performs significantly well under favorable weather but suffers miss detections and false positives due to atmospheric scattering caused by fog particles. Existing deep object detection techniques often exhibit a high degree of accuracy. The drawback is being sluggish at object detection in fog. Object detection methods with fast detection speed have been obtained using deep learning at the expense of accuracy. The problem of the lack of balance between detection speed and accuracy in fog persist. This paper presents an improved YOLOv5-based multi-sensor fusion network that combines radar's object detection with a camera image bounding box. We transformed radar detection by mapping the radar detections into a two-dimensional image coordinate and projected the resultant radar image on the camera image. Using the attention mechanism, we emphasized and improved important feature representation used for object detection while reducing high-level feature information loss. We trained and tested our multi-sensor fusion network on clear and multi-fog weather datasets obtained from the CARLA simulator. Our result shows that the proposed method significantly enhances the detection of distant and small objects. Our small CR-YOLOnet model best strikes a balance between accuracy and speed with an accuracy of 0.849 at 69 fps.

The traditional sequent peak algorithm (SPA) was used to assess the reservoir volume (VR) for comparison with deficit volume, DT, (subscript T representing the return period) obtained from the drought magnitude (DM) based method with draft level set at the mean annual flow on 15 rivers across Canada. At an annual scale, the SPA based estimates were found to be larger with an average of nearly 70% compared to DM based estimates. To ramp up DM based estimates to be in parity with SPA based values, the analysis was carried out through the counting and the analytical procedures involving only the annual SHI (standardized hydrological index, i.e. standardized values of annual flows) sequences. It was found that MA2 or MA3 (moving average of 2 or 3 consecutive values) of SHI sequences were required to match the counted values of DT to VR. Further, the inclusion of mean, as well as the variance of the drought intensity in the analytical procedure, with aforesaid smoothing led DT comparable to VR. The distinctive point in the DM based method is that no assumption is necessary such as the reservoir being full at the beginning of the analysis - as is the case with SPA.

Contact with nature is suggested to be as an important resource for human health and well-being. Stress alleviation and reduction of mental fatigue are main instances of improvement of human health and well-being in contact with nature. Although the term restoration experience in contact with nature has been used repeatedly for concepts of stress alleviation and reduction of mental fatigue, these refer to different sets of theories and research studies. In this paper, by reviewing the nature-health related theories of Biophilia, Stress Recovery Theory, Attention Restoration Theory and Supportive Environment Theory (SET), the similarities and differences between these two concepts have been discussed. Moreover, the measurement approach instruments based on different study methodologies and aims are provided. This paper could provide information for future studies with focus on factors to be of benefit for restoration experience in design and planning of natural landscape environments with the aim of health and well-being development.

In recent years, the analysis of macro- and micro-expression has drawn the attention of researchers since they provide visual cues of an individual's emotions for a broad range of potential applications such as lie detection and criminal detection. In this paper, we address the challenge of spotting facial macro- and micro-expression from videos and present compelling results by using a deep learning approach to analyze the optical flow features. Different from other deep learning approaches that are mainly based on Convolutional Neural Networks (CNNs), we propose a Transformer-based deep learning approach that predicts a score indicating the probability of a frame being within an expression interval. In contrast to other Transformer-based models that achieve high performance by being pre-trained on large datasets, our deep learning model, called SL-Swin, which incorporates Shifted Patch Tokenization and Locality Self-Attention into the backbone network Swin Transformer, effectively spots macro- and micro-expressions by being trained from scratch on small-size expression datasets. Our evaluation outcomes surpass the MEGC 2022 spotting baseline result, obtaining an overall F1-score of 0.1366. Additionally, our approach also performs well in the MEGC 2021 spotting task with an overall F1-score of 0.1824 and 0.1357 on CAS(ME)^2 and SAMM Long Videos, respectively. The code is publicly available on GitHub (https://github.com/eddiehe99/pytorch-expression-spotting).

Due to the special characteristics of the shooting distance and angle of remote sensing satellites, the pixel area ratio of ship targets is small and the feature expression is insufficient, which leads to unsatisfactory ship detection performance and even situations such as missed detection and false detection. In this study, we propose an improved-YOLOv5 algorithm. The improvement strategies mainly include: (1) Add the Convolutional Block Attention Module (CBAM) into the Backbone to enhance the extraction of target-adaptive optimal features; (2) Introduce cross-layer connection channel and lightweight GSConv structure into the Neck to achieve higher-level multi-scale feature fusion and reduce the number of model parameters; (3) The Wise-IoU loss function is used to cal-culate the localization loss in the Output, and assign reasonable gradient gains to cope with dif-ferences in image quality. In addition, during the preprocessing stage of experimental data, a me-dian and bilateral filter method is used for noise reduction to reduce interference from ripples and waves and highlight the information of ship features. The experimental results show that Im-proved-YOLOv5 has a significant improvement in recognition accuracy compared to various mainstream target detection algorithms; Compared to the original YOLOv5s, the mean Average Precision (mAP) has improved by 3.2% and the Frames Per Second (FPN) has accelerated by 8.7%.

In the process of climate change mitigation and adaptation, the focus on climate risks driven by transition innovations has received widespread attention from governments, regulators and investors. Based on the identification of low-carbon patents through the Green and Low-carbon Technology Inventory in China, we construct low-carbon innovation measurements of listed firms in China from 2015 to 2021 then an empirical model is used to investigate the impact and mechanism of climate transition innovations on the default risk. The baseline regression model shows that low-carbon innovations can mitigate the default risk of firms. Moreover, our findings are supported by instrumental variable regressions that use time costs of innovation. In addition, the mediation effect demonstrates that investor attention, total factor productivity, and technology spillovers are paths for transition innovations to affect the default risk. This study reveals that low-carbon technological advances have a positive effect on climate transition risks. Further, it provides empirical evidence for listed companies to develop low-carbon innovations.

Speech Emotion Recognition (SER) is an interesting and difficult problem to handle. In this paper, we deal with it through the implementation of deep learning networks. We have designed and implemented six different deep learning networks, a Deep Belief Network (DBN), a simple deep neural network (SDNN), a LSTM network (LSTM), a LSTM network with the addition of an attention mechanism (LSTM-ATN), a Convolutional neural network (CNN), and a Convolutional neural network with the addition of an attention mechanism (CNN-ATN), having in mind, apart from solving the SER problem, to test the impact of attention mechanism to the results. Dropout and Batch Normalization techniques are also used to improve the generalization ability (prevention of overfitting) of the models as well as to speed up the training process. The Surrey Audio-Visual Expressed Emotion database (SAVEE), and the Ryerson Audio-Visual Database (RAVDESS) database were used for training and evaluation of our models. The results showed that networks with the addition of the attention mechanism did better than the others. Furthermore, they showed that CNN-ATN was the best among tested networks, achieving an accuracy of 74% for the SAVEE and 77% for the RAVDESS dataset, and exceeded existing state-of-the-art systems for the same datasets.

Attention deficit hyperactivity disorder (ADHD) results in various functioning impairments in children’s lives and families. Parents of children with ADHD report high levels of parenting stress, low levels of parental self-efficacy and use of more authoritarian and/or permissive parenting practices than parents of typically developing children. Intervention programs need to address both children’s and parents’ needs and multimodal intervention programs could cover this demand. The aim of this study was to examine the efficacy of “Child ViReal Support Program” - a multi-level evidence-based comprehensive program - on parenting stress, parental self-efficacy, parenting practices, and the core symptoms of children’s ADHD. Families with a child diagnosed with ADHD (n = 16) were randomly allocated to two groups (PC and CP; P = parent training, C = child training) and a cross-over design was utilized. Participating parents completed, in four different times during the study, the Parenting Stress Index-Short Form, the Parenting Sense of Competence Scale, the Parenting Styles and Dimensions Questionnaire-Short Version and the parent form of the ADHD Rating Scale-IV. Parents from both groups, after their participation in the parent training, demonstrated reduced parenting stress, enhanced parental self-efficacy and increased the employ of democratic parenting practices. More than that, they reported decreased levels of inattention and impulsivity/hyperactivity for their children. Evidence-based multi-level intervention programs could produce positive effects in parents and children by incorporating effective methods and tools in accordance with the needs and the demands of the family context.

Transformer models have emerged as frontrunners in the field of natural language processing, primarily due to their adept use of self-attention mechanisms to grasp the semantic linkages between words in sequences. Despite their strengths, these models often face challenges in single-task learning scenarios, particularly when it comes to delivering top-notch performance and crafting strong latent feature representations. This challenge is more pronounced in the context of smaller datasets and is particularly acute for under-resourced languages such as Arabic. In light of these challenges, this study introduces a novel methodology for text classification of Arabic texts. This method harnesses the newly developed Reverse Positional Encoding (RPE) technique. It adopts inductive-transfer learning (ITL) framework combined with a switching self-attention shared encoder, thereby increasing the model's adaptability and improving its sentence representation accuracy. The integration of Mixture of Experts (MoE) and Reverse Positional Encoding (RPE) techniques empowers the model to process longer sequences more effectively. This enhancement is notably beneficial for Arabic text classification, adeptly supporting both the intricate five-point and the simpler ternary classification tasks. The empirical evidence points to its outstanding performance, achieving accuracy rates of 87.20% for the HARD dataset, 72.17% for the BRAD dataset, and 86.89% for the LABR dataset, as evidenced by the assessments conducted on these datasets.

Document classification is one of the most critical steps in the document analysis pipeline. There are two types of approaches for document classification, known as image-based and multimodal approaches. The image-based document classification approaches are solely based on the inherent visual cues of the document images. In contrast, the multimodal approach co-learns the visual and textual features, and it has proved to be more effective. Nonetheless, these approaches require a huge amount of data. This paper presents a novel approach for document classification that works with a small amount of data and outperforms other approaches. The proposed approach incorporates a hierarchical attention network(HAN) for the textual stream and the EfficientNet-B0 for the image stream. The hierarchical attention network in the textual stream uses the dynamic word embedding through fine-tuned BERT. HAN incorporates both the word level and sentence level features. While the earlier approaches rely on training on a large corpus (RVL-CDIP), we show that our approach works with a small amount of data (Tobacco-3482). To this end, we trained the neural network at Tobacco-3428 from scratch. Thereby, we outperform state-of-the-art by obtaining an accuracy of 90.3%. This results in a relative error reduction rate of 7.9%.

Human violence recognition is an area of great interest in the scientific community, given its broad spectrum of applications, especially in video surveillance systems, since detecting violence in real-time could prevent criminal acts and save lives. Despite the number of existing proposals and research, most focus on the precision of results, leaving aside efficiency and its practical implementation. Thus, this work proposes a model that is effective and efficient in recognizing human violence in real-time. The proposed model consists of three modules: a first module called Spatial Motion Extractor (SME), in charge of extracting regions of interest from a frame; a second module called Short Temporal Extractor (STC), whose function is to extract temporal characteristics of rapid movements, finally the Global Temporal Extractor (GET) module, responsible for identifying long-lasting temporal features and fine-tuning the model. The proposal was evaluated regarding efficiency, effectiveness, and ability to operate in real-time. The results obtained on Hockey, Movies, and RWF-2000 datasets demonstrated that this approach is highly efficient compared to other alternatives. A VioPeru dataset was created to validate real-time applicability with violent and non-violent videos captured by real video surveillance cameras in Peru. The effectiveness results in this dataset outperformed the best existing proposal. Therefore, our proposal has contributions in efficiency, effectiveness, and real-time.

The challenge of campus design, like other aspects of contemporary environmental design, reveals a serious problem in education and practice. The foundational design theories of a century ago have been exposed as an obsolete way of thinking about cities, human nature, biological nature, and even the nature of mathematical and physical structures. Yet in practice, these discredited models persist, obscured by new theoretical language and extravagant “neoplastic” forms, but embodying persistent though untested ideologies and driven by systems inertia. This paper considers the campus design typology (including business campuses, commercial districts, hospitals, and schools) as a design paradigm for pedestrian public space with implications for human flourishing and well-being. We propose a specific human-oriented design method to encourage the well-being of occupants and improve the outcomes for creative development, education, and health.

Craters are the most prominent geomorphological features on the surface of celestial bodies, which is playing a crucial role in studying the formation and evolution of celestial bodies, as well as in landing and planning for surface exploration. Currently, the main automatic crater detection models and datasets focus on the detection of large and medium craters. In this paper, we created 23 small lunar crater datasets for model training based on the Chang’E-2 (CE-2) DOM, DEM, Slope, and integrated data with 7 kinds of visualization stretching methods. And then, we proposed the YOLO-Crater model for Lunar and Martian small crater detection by replacing EioU and VariFocal loss to solve the crater samples imbalance problem and introducing a CBAM attention mechanism to mitigate interference from the complex extraterrestrial environment. The results show that the accuracy (P = 87.86%, R = 66.04%, and F1 = 75.41%) of the Lunar YOLO-Crater model based on the DOM-MMS (Maximum-Minimum Stretching) dataset is the highest and better than that of YOLOX model. And the Martian YOLO-Crater, trained by the Martian dataset from the 2022 GeoAI Martian Challenge, achieves good performance with P = 88.37%, R = 69.25%, and F1 = 77.65%. It indicates that the YOLO-Crater model has a strong transferability and generalization capability, which can be applied to detect small craters on the Moon and other celestial bodies.

The unstable international economic situation is reflected in the supply chain stress, lack or increased cost of some raw materials, fuel or semi-finished products is forcing organizations to perform new optimization initiatives in the utilization of their equipment and assets pointed to obtain the maximum value from them, while maintaining and even improving the quality of their products. The achievement of these objectives involves the reduction or minimization of equipment downtime to maintain the advantage over their competitors and ensure the organization's competitiveness. The intelligent maintenance system (IMS) provides adequate support for decision-making related to equipment maintenance, since poor maintenance results in unplanned stoppages, with the consequent additional cost and increased customer dissatisfaction, and an over-maintenance can result in an additional labor cost, time and the replacement of parts that are in good conditions. The utilization of new tools and technologies introduced by Industry 4.0 offers multiple opportunities for enhancement through communication and computerized data processing, aiming to improve the maintainability of a hydrogen compressor using neural networks based on attention mechanisms combined with linear regression.

Automated extraction of buildings from earth observation (EO) data is important for various applications, including updating of maps, risk assessment, urban planning, policy making. Combining data from different sensors such as high-resolution multispectral (HRI) and light detection and ranging (LiDAR) has shown great potential in building extraction. Deep learning (DL) is increasingly used in multimodal data fusion and urban object extraction. However, DL-based multimodal fusion networks may underperform due to insufficient learning of “joint features” from multiple sources and oversimplified approaches to fusing multimodal features. Recently, an hybrid attention-aware fusion network (HAFNet) has been proposed for building extraction from a dataset including co-located Very-High-Resolution (VHR) optical images and Light Detection And Ranging (LiDAR) joint data. The system reported good performances thanks to the adaptivity of the attention mechanism to the features of the information content of the three streams but suffered from model overparametrization, which inevitably leads to long training times and heavy computational load. In this paper the authors propose a restructuring of the scheme, which involved replacing VGG-16-like encoders with the recently proposed EfficientNet, whose advantages counteract exactly the issues found with the HAFNet scheme. The novel configuration was tested on multiple benchmark datasets, reporting great improvements in terms of processing times, and also in terms of accuracy. The new scheme, called HAFNetE (HAFNet with EfficientNet integration), appears indeed capable of achieving good results with less parameters, translating into better computational efficiency. Based on these findings, we can conclude that, given the current advancements in single-thread schemes, the classical multi-thread HAFNet scheme could be effectively transformed by the HAFNetE scheme by replacing VGG-16 with EfficientNet blocks on each single thread. The remarkable reduction achieved in computational requirements moves the system one step closer to on-board implementation in a possible, future “urban mapping” satellite constellation.

The technical capabilities of modern Industry 4.0 and Industry 5.0 are rather vast and growing exponentially daily. The present-day Industrial Internet of Things (IIoT) combines manifold underlying technologies that require real-time interconnections and communications among heterogeneous devices. Smart cities are established with sophisticated designs and control for seamless Machine-to-Machine (M2M) communication to optimize resources, costs, performances, and energy distribution. All the sensory devices within a building interact to maintain a sustainable climate for residents and intuitively optimize the energy distribution to optimize energy production. However, it encompasses quite a few challenges for devices that lack compatible and interoperable designs. Conventional solutions are restricted to limited domains or rely on engineers to design and deploy translators for each pair of ontologies. This is a costly process in terms of engineering efforts and computational resources. The issue persists that a new device with a different ontology must be integrated into an existing IoT network. We propose a self-learning model that can determine the taxonomy of devices given their ontology meta-data and structural information. The model finds matches between two distinct ontologies using the Natural Language Processing (NLP) approach for learning linguistic contexts. Then, by visualizing the ontology network as a knowledge graph, it is possible to learn the structure of the meta-data and understand the device's message formulation. Finally, it can align entities of both ontology graphs similar in context and structure. Furthermore, the model performs dynamic M2M translation without requiring extra engineering or hardware efforts.

One of the most rapidly advancing areas of deep learning research aims at creating models that learn to disentangle the latent factors of variation from a data distribution. However, modeling joint probability mass functions is usually prohibitive, which motivates the use of conditional models assuming that some information is given as input. In the domain of numerical cognition, deep learning architectures have successfully demonstrated that approximate numerosity representations can emerge in multi-layer networks that build latent representations of a set of images with a varying number of items. However, existing models have focused on tasks requiring to conditionally estimate numerosity information from a given image. Here we focus on a set of much more challenging tasks, which require to conditionally generate synthetic images containing a given number of items. We show that attention-based architectures operating at the pixel level can learn to produce well-formed images approximately containing a specific number of items, even when the target numerosity was not present in the training distribution.

This paper provides a review of an emerging field in the food processing sector, referring to efficient and safe food supply chains, ’from farm to fork’, as enabled by Artificial Intelligence (AI). Recent advances in machine and deep learning are used for effective food production, energy management and food labeling. Appropriate deep neural architectures are adopted and used for this purpose, including Fully Convolutional Networks, Long Short-Term Memories and Recurrent Neural Networks, Auto-Encoders and Attention mechanisms, Latent Variable extraction and clustering, as well as Domain Adaptation. Three experimental studies are presented, illustrating the ability of these AI methodologies to produce state-of-the-art performance in the whole food supply chain. In particular, these concern: (i) predicting plant growth and tomato yield in greenhouses, thus matching food production to market needs and reducing food waste or food unavailability; (ii) optimizing energy consumption across large networks of food retail refrigeration systems, through optimal selection of systems that can get shut-down and through prediction of the respective food de-freezing times, during peaks of power demand load; (iii) optical recognition and verification of food consumption expiry date in automatic inspection of retail packaged food, thus ensuring safety of food and people’s health.

Recent studies have proven that synthetic aperture radar (SAR) automatic target recognition (ATR) models based on deep neural networks (DNN) are vulnerable to adversarial examples. However, existing attacks are easily failed in the case where adversarial perturbations cannot be fully fed to victim models. We call this situation perturbation offset. Moreover, since background clutter takes up most of the areas in SAR images and has low relevance to recognition results, fooling models with global perturbations is quite inefficient. This paper proposes a semi-whitebox attack network, called Universal Local Adversarial Network (ULAN), to generate universal adversarial perturbations (UAP) for the target regions of SAR images. In the proposed network, we calculate the model’s attention heatmaps through layer-wise relevance propagation (LRP), which is used to locate the target regions of SAR images that have high relevance to recognition results. In particular, we utilize a generator based on the U-Net to learn the mapping from noise to UAPs and craft adversarial examples by adding the generated local perturbations to target regions. Experiments indicate that the proposed method fundamentally prevents perturbation offset and achieves comparable attack performance to conventional global UAPs by perturbing only a quarter or less of SAR image areas.

Predatory research field in science is when an emerging scientific topic destroys current topics and characterizes a main scientific change. Predatory research field can be a basic driver of scientific and technological change that generates a 'creative destruction' in science and society in contexts of knowledge-based competition and rapid changes. The prediction of proposed theory of predatory research fields is that it destroys with a fast growth other research fields. The theoretical approach is tested here in research fields of large language models (LLM) by analyzing the transformers (a deep learning architecture based on the multi-head attention mechanism) proposed in 2017 and from November 2022 started main applications in generative artificial intelligence with innovations of BERT, ChatGPT, Microsoft Copilot (launched on February 7, 2023) and other natural language processing tools driven by AI technology for engaging conversations, gain insights, automate tasks, etc., etc. Statistical evidence suggests that growth rate of transformer technologies is 80.58%, a high level compared to all other research fields in machine learning (having a growth rate of 13.83%). Moreover, predatory research field of transformers has a destructive power such that all other domains in LLM from 2021 to 2023 have a general reduction of scientific growth. The impact of transformers is much more drastic of previous radical technologies such as CNN having a temporal growth rate of 0.16%, lower than 0.38% by transformers, ceteris paribus. These analysis reveals that transformers have characteristics to generate a radical scientific and technological change in a not-too-distant future. Overall, then, the study suggests that predatory research field on emerging topics and technologies can generate path-breaking innovations and the examination here can clarify the essential elements of the science dynamics for a better theory of scientific and technological change, providing also main implications for knowledge policy to support promising research fields and technologies to guide economic and social change.

Understanding brain function is one of the most important problems in human history. At present, there is no concrete theory for how the brain works. Here, a theory is presented that provides a detailed mechanistic biological account of the brain's capacities, including motor control, functional states, language, and thinking. Brain function is managed by a well-defined response (R) process that is generally similar to the process underlying the immune system. The R process is strongly reflected in the brain's anatomy, physiology, and external interactions. Different R process stages are supported by distinct excitatory networks located in different cortical layers, hippocampal fields, and bagal ganglia paths, by distinct coordination networks comprised of GABAergic interneurons, and by distinct molecular agents. The roles of norepinephrine, serotonin, dopamine and acetylcholine is to promote the alert, planning, goal-setting and execution R process modes, respectively. Opioids and oxytocin promote termination by success, failure, fight or run, while glucocorticoids and cannabinoids suppress acute responses to protect cells. The R process has two instances occurring at different time scales. The millisecond-scale Quax process implements the execution of hierarchical sequences of movements and thoughts, in which the selection of the next action is determined via interaction between top-down predictions and sensory inputs. The slower Need process controls the satisfaction of internal and external needs. The theory differs from the existing standard accounts in many of the major topics (e.g., the basal ganglia, dopamine, language), and shows how cognition results from biological processes.