Policy iteration approximate dynamic programming using Volterra series based actor.
Wentao GuoJennie SiFeng LiuShengwei MeiPublished in: IJCNN (2014)
Keyphrases
- approximate dynamic programming
- policy iteration
- least squares
- markov decision processes
- reinforcement learning
- model free
- optimal policy
- factored mdps
- fixed point
- temporal difference
- infinite horizon
- markov decision process
- actor critic
- linear program
- finite state
- dynamic programming
- average reward
- markov decision problems
- optimal control
- state space
- convergence rate
- linear programming
- average cost
- step size
- control policy
- multiresolution
- decision problems
- action space
- multistage
- random variables
- optical flow
- lower bound