Unified Policy Optimization for Continuous-action Reinforcement Learning in Non-stationary Tasks and Games.
Rong-Jun QinFan-Ming LuoHong QianYang YuPublished in: CoRR (2022)
Keyphrases
- non stationary
- policy search
- continuous action
- reinforcement learning
- continuous state
- action space
- optimal policy
- reinforcement learning algorithms
- state space
- transfer learning
- partially observable markov decision processes
- dynamic programming
- random fields
- markov decision process
- change point detection
- policy gradient
- reward function
- markov decision processes
- control policies
- stochastic processes
- function approximators
- empirical mode decomposition
- state action
- learning algorithm
- object detection
- reinforcement learning methods
- model free
- game theory