Explore the foundational concepts of statistical considerations in reinforcement learning in this comprehensive lecture from the Theory of Reinforcement Learning Boot Camp. Delve into topics such as inverse reinforcement learning, high-stakes problems, and the Max of Means Problem. Examine unique maximizers, simulations, and uncertainty in reinforcement learning contexts. Investigate local parameter asymptotics, natural estimators, and the moving parameter asymptotic framework. Gain insights into projection intervals and conservative estimators as presented by Eric Laber from North Carolina State University in this 1-hour and 14-minute talk hosted by the Simons Institute.
Statistical Considerations in Reinforcement Learning - Part 1a
Overview
Syllabus
Introduction
Background
Example
Inverse Reinforcement Learning
Highstakes problems
Notes on context
Max of Means Problem
Unique Maximizer
Simulation
Uncertainty
Local parameter asymptotics
Natural estimator
Moving parameter asymptotic framework
Projection interval
Conservative estimator
Taught by
Simons Institute
