Rates of Convergence of Performance Gradient Estimates Using Function Approximation and Bias in Reinforcement Learning

Grudic, Gregory Z., Ungar, Lyle H.

–Neural Information Processing Systems 

We address two open theoretical questions in Policy Gradient Reinforcement Learning. The first concerns the efficacy of using function approximation to represent the state action value function,.

Neural Information Processing Systems

Dec-31-2002

Conferences    PDF

Add feedback

Duplicate Docs Excel Report

Title
Rates of Convergence of Performance Gradient Estimates Using Function Approximation and Bias in Reinforcement Learning
Rates of Convergence of Performance Gradient Estimates Using Function Approximation and Bias in Reinforcement Learning

Similar Docs  Excel Report  more

TitleSimilaritySource
None found