Unified Policy Optimization for Robust Reinforcement Learning.
Zichuan LinLi ZhaoJiang BianTao QinGuangwen YangPublished in: ACML (2019)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- global optimization
- model free
- approximate dynamic programming
- control policy
- policy iteration
- markov decision process
- robust optimization
- action selection
- partially observable environments
- learning algorithm
- markov decision problems
- reward function
- function approximation
- computationally efficient
- markov chain
- markov decision processes
- optimization algorithm
- sufficient conditions
- control policies
- optimization problems
- least squares
- state space