Formisano, A.; Tucci, M. Machine Learning Approaches for Inverse Problems and Optimal Design in Electromagnetism. Electronics2024, 13, 1167.
Formisano, A.; Tucci, M. Machine Learning Approaches for Inverse Problems and Optimal Design in Electromagnetism. Electronics 2024, 13, 1167.
Formisano, A.; Tucci, M. Machine Learning Approaches for Inverse Problems and Optimal Design in Electromagnetism. Electronics2024, 13, 1167.
Formisano, A.; Tucci, M. Machine Learning Approaches for Inverse Problems and Optimal Design in Electromagnetism. Electronics 2024, 13, 1167.
Abstract
The spread of high-performance personal computers, frequently equipped with powerful Graphic Processing Units (GPU), raised the interest on a set of techniques able to extract models of electromagnetic phenomena (and devices) directly from available examples of the desired behavior. Such approaches are collectively referred to as Machine Learning (ML). A typical representative ML approach is the so called “Neural Network” (NN). Using such data-driven models allows evaluating the output in a much shorter time when a theoretical model is available, or allows predicting the behavior of the systems and devices when no theoretical model is available. With reference to a simple yet representative benchmark electromagnetic problem, some of the possibilities and the pitfalls of the use of NN for the interpretation of measurements (inverse problem) or to obtain the required measurements (optimal design problem) are discussed. The investigated aspects include the choice of the NN model; the generation of the dataset(s); the selection of hyperparameters (hidden layers, training paradigm). Finally, the capabilities in the handling of ill-posed problems are critically revised.
Keywords
Machine Learning; Magnetic Field Analysis; Optimal Design, Inverse Problems.
Subject
Engineering, Electrical and Electronic Engineering
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
