Empirical policy iteration for approximate dynamic programming.
William B. HaskellRahul JainDileep M. KalathilPublished in: CDC (2014)
Keyphrases
- approximate dynamic programming
- policy iteration
- markov decision processes
- model free
- reinforcement learning
- optimal policy
- least squares
- fixed point
- markov decision process
- factored mdps
- finite state
- average reward
- dynamic programming
- linear program
- temporal difference
- linear programming
- infinite horizon
- state space
- convergence rate
- actor critic
- machine learning
- average cost
- step size
- function approximation
- markov decision problems
- optimal control
- partially observable
- markov chain
- search algorithm
- image sequences
- data mining