Enabling Surrogate-Assisted Evolutionary Reinforcement Learning via Policy Embedding.
Lan TangXiaxi LiJinyuan ZhangGuiying LiPeng YangKe TangPublished in: BIC-TA (2022)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- action selection
- markov decision process
- markov decision processes
- state space
- actor critic
- control policy
- partially observable environments
- policy iteration
- function approximators
- markov decision problems
- policy gradient
- function approximation
- policy evaluation
- reinforcement learning problems
- control policies
- action space
- partially observable
- evolutionary optimization
- approximate dynamic programming
- state and action spaces
- reward function
- genetic algorithm
- dynamic programming
- decision problems
- reinforcement learning algorithms
- finite state
- vector space
- infinite horizon
- rl algorithms
- partially observable domains
- evolutionary computation
- partially observable markov decision processes
- inverse reinforcement learning
- learning problems
- state action
- transition model
- learning process
- model free
- temporal difference
- continuous state spaces
- sufficient conditions
- optimal control
- machine learning
- watermarking algorithm
- state dependent
- reinforcement learning methods
- average reward
- learning algorithm
- transfer learning
- markov chain
- continuous state
- evolutionary process
- nonlinear dimensionality reduction