Statistical Physics of Learning

This lecture provides a brief review of 40 years of statistical physics of learning, focusing on the relationship between the structure of neural networks and the results obtained from the statistical physics of disordered systems. It covers topics such as learning algorithms with optimal stability, information storage and retrieval in spin-glass models of neural networks, and the mean field approach to Bayes learning in feed-forward neural networks. The lecture delves into the theoretical foundations and practical implications of statistical mechanics in understanding the learning process in neural networks.