Convergence of Least Squares Temporal Difference Methods Under General Conditions.
Huizhen YuPublished in: ICML (2010)
Keyphrases
- general conditions
- least squares
- temporal difference methods
- policy evaluation
- temporal difference
- function approximation
- convergence rate
- optical flow
- evolutionary methods
- reinforcement learning
- convergence speed
- reinforcement learning problems
- policy search
- td learning
- policy iteration
- variance reduction
- function approximators
- machine learning
- active learning
- genetic algorithm