Overview
Explore Bayesian Reinforcement Learning in this comprehensive lecture covering key concepts such as inaccurate models, planning, sampling, value iteration, posterior belief, and the exploration-exploitation tradeoff. Delve into both offline and online RL approaches while gaining insights into the benefits of Bayesian methods in reinforcement learning. Learn from Pascal Poupart as he provides a thorough introduction and recap of RL fundamentals before diving deep into Bayesian RL techniques.
Syllabus
Introduction
Recap
Inaccurate Models
Benefits
RL Recap
Bayesian RL
Planning
Sampling
Value iteration
posterior belief
exploration exploitation tradeoff
offline and online RL
Taught by
Pascal Poupart