Policy Decomposition: Approximate Optimal Control with Suboptimality Estimates.
Ashwin KhadkeHartmut GeyerPublished in: HUMANOIDS (2021)
Keyphrases
- optimal control
- infinite horizon
- stochastic control
- dynamic programming
- actor critic
- average cost
- control problems
- policy gradient
- policy iteration
- feedback control
- reinforcement learning
- optimal policy
- risk sensitive
- control strategy
- policy evaluation
- class of nonlinear systems
- lyapunov function
- optimal control problems
- decomposition method
- markov decision process
- brownian motion
- finite horizon
- markov decision problems
- markov decision processes
- stochastic demand
- neural network
- control law