Joint Action Representation and Prioritized Experience Replay for Reinforcement Learning in Large Discrete Action Spaces.
Xueyu WeiWei XueWei ZhaoYuanxia ShenGaohang YuPublished in: ICMLSC (2023)
Keyphrases
- action space
- reinforcement learning
- continuous state
- joint action
- state space
- continuous action
- continuous state spaces
- markov decision processes
- state and action spaces
- policy search
- real valued
- stochastic processes
- multiagent reinforcement learning
- markov decision process
- reinforcement learning algorithms
- state action
- action selection
- model free
- optimal policy
- partially observable
- finite state
- policy iteration
- machine learning
- temporal difference
- optimal control
- function approximation
- learning algorithm