36-734 Topics in Learning Theory II

6 units, Carnegie Mellon University

This mini explores bandit algorithms and online learning, foundational to reinforcement learning and the fine-tuning of large language models. The main reference is Lattimore and Szepesvári, Bandit Algorithms (Cambridge University Press, 2020). Additional readings may include Six Lectures on Linearized Neural Networks (Misiakiewicz & Montanari), Reinforcement Learning: Theory and Algorithms (Agarwal et al.), and A Mathematical Perspective on Transformers (Geshkovski et al.). This sequence is intended for advanced PhD students in statistics and machine learning seeking a rigorous understanding of the statistical foundations of deep and online learning.

Prerequisites: none listed

Offered: Spring 2026

Courseweave needs JavaScript to show sections, workloads and planning tools.