Statistics 690B - Deep Learning

This course provides a rigorous and comprehensive introduction to the theoretical foundations, algorithmic principles, and advanced concepts in deep learning. The course is designed to equip students with both the mathematical underpinnings and practical insights necessary to conduct cutting-edge research in the field. We will explore key topics such as neural network architectures, optimization techniques, probabilistic deep learning, generative models, and modern advancements in the discipline, such as transformers, graphical neural networks, generative adversarial networks, normalizing flows, autoencoders, and diffusion models. Emphasis will be placed on developing a deep conceptual understanding, critical analysis of research papers, and the ability to formulate and solve novel problems in deep learning. Students will engage in theoretical derivations, hands-on implementations, and research-oriented discussions to prepare them for original contributions in academia or industry.

STATISTC 525/625, 535 & 607&08