NYU Deep Learning - Fall 2022

Learn deep learning fundamentals through this comprehensive lecture series covering neural network inference, energy-based models, classification techniques, and advanced topics like latent variable models. Explore the mathematical foundations of backpropagation and contrastive learning while gaining hands-on experience with PyTorch implementation through practical 5-step training code examples. Dive into energy perspectives on classification problems and understand how they connect to broader machine learning concepts. Master latent variable energy-based models (LV-EBMs) for both inference and training scenarios, progressing from basic concepts like K-means clustering and sparse coding to more sophisticated approaches including target propagation and autoencoders. Build a solid theoretical foundation while developing practical skills in implementing and training deep learning models using modern frameworks.