Model-Based Least-Squares Policy Evaluation.
Fletcher LuDale SchuurmansPublished in: Canadian Conference on AI (2003)
Keyphrases
- policy evaluation
- least squares
- model free
- matrix inversion
- policy iteration
- linear regression
- bellman residual
- parameter estimation
- reinforcement learning
- monte carlo
- markov decision processes
- optical flow
- temporal difference
- function approximation
- dynamic programming
- semi parametric
- variance reduction
- linear model
- support vector
- objective function
- image sequences