Advanced Deep Learning

Catalog description

Deep learning with artificial neural nets underpins all modern AI, such as large language models and image and video generators. This course presents multiple advanced topics in some depth, including advanced convolutional nets, generative AI and reinforcement learning (RL). Generative models include transformer-based autoregressive language models, generative adversarial networks (GANs), variational autoencoders (VAEs) and diffusion models. Reinforcement learning topics include Gymnasium (OpenAI Gym), temporal difference learning, deep Q learning, RL libraries and world models. Emphasizes building models in Python using PyTorch. Also covers implementing workflows using libraries and transfer learning with pre-trained models, AI agents and retrieval augmented generation (RAG). Assumes familiarity with Python and previous experience with machine learning and PyTorch.