Amirkabir University of Technology AUT Journal of Electrical Engineering 2588-2910 56 3 2024 07 01 A New Physics-Inspired Discriminative Classifier 495 502 5405 10.22060/eej.2024.22694.5557 EN Mostafa Monemizadeh Department of Electrical and Computer Engineering, University of Neyshabur, Neyshabur, Iran 0000-0001-8521-8436 Seyed Rouhollah Samareh Hashemi Fiber Optics Group, Institute of Science and High Technology and Environmental Sciences, Kerman, Iran 0000-0002-1450-7692 Mohsen Sheikh-Hosseini Department of Computer and Information Technology, Institute of Science and High Technology and Environmental Sciences, Kerman, Iran 0000-0002-5959-7839 Hamed Fehri Faculty of Engineering, University of Zabol, Zabol, Iran 0000-0001-8521-8436 Journal Article 2023 09 22 Concepts and laws of physics have been a valuable source of inspiration for engineers to overcome human challenges and problems. Classification is an important example of such problems that play a major role in various fields of engineering sciences. It is shown that discriminative classifiers tend to outperform their generative counterparts, especially in the presence of sufficient labeled training data. In this paper, we present a new physics-inspired discriminative classification method<strong> </strong>using minimum potential lines.  To do this, we first consider two groups of fixed point charges (as two classes of data) and a movable classifier line between them. Then, we find a stable position for the classifier line by minimizing the total potential integral on the classifier line due to the two groups of point charges. Surprisingly, it will be shown that the obtained classifier is actually an uncertainty-based classifier that minimizes the total uncertainty of the classifier line. Experimental results show the effectiveness of the proposed approach. https://eej.aut.ac.ir/article_5405_ca6ab34959489659f8c3776aaf1f8efd.pdf