3.0

Average Course Rating

Recent advancement in language models have given rise to a new generation of autonomous reasoning agents capable of executing valuable and challenging tasks. This advanced graduate course covers the emerging space of agentic systems with an emphasis on “safely confident reasoning". We will jointly study the design of agents and the development of the foundational models that power these systems. We will discuss topics integral to real-world agent deployments, such as efficiency and security concerns, uncertainty quantification and explainability for accountable decision making, indexing and retrieval tools that power agent memory. We conclude by discussing future directions for this emerging technology. Required Course Background: Machine Learning, ML: Deep Learning, ML: Learning Theory, ML: Reinforcement Learning, ML: Artificial System Design and Development, NLP: Self-Supervised Models, or equivalent (grad course in machine learning).