Self-Supervised Learning and Variational Inference

Self-Supervised Learning and Variational Inference

Dive into a comprehensive lecture on self-supervised learning and variational inference, presented by renowned speaker Yann LeCun. Explore key concepts including GANs, sparse modeling, amortized inference, and convolutional sparse modeling with group sparsity. Gain insights into discriminant recurrent sparse autoencoders and various self-supervised learning techniques. Examine regularization through temporal consistency and delve into Variational Autoencoders (VAEs), covering both intuitive and probabilistic variational approximation-based interpretations. Enhance your understanding of advanced machine learning concepts in this nearly two-hour session, part of a broader deep learning course series.