Value and Policy Iteration in Optimal Control and Adaptive Dynamic Programming.
Dimitri P. BertsekasPublished in: CoRR (2015)
Keyphrases
- optimal control
- policy iteration
- dynamic programming
- infinite horizon
- markov decision processes
- control problems
- actor critic
- optimal policy
- approximate dynamic programming
- control strategy
- reinforcement learning
- average cost
- markov decision problems
- multistage
- state space
- linear programming
- average reward
- optimal control problems
- partially observable