A Policy-Guided Imitation Approach for Offline Reinforcement Learning.
Haoran XuLi JiangJianxiong LiXianyuan ZhanPublished in: NeurIPS (2022)
Keyphrases
- reinforcement learning
- optimal policy
- policy search
- markov decision process
- action selection
- partially observable
- action space
- policy iteration
- state space
- reward function
- actor critic
- control policies
- markov decision problems
- function approximators
- rl algorithms
- function approximation
- partially observable environments
- approximate dynamic programming
- reinforcement learning problems
- reinforcement learning algorithms
- markov decision processes
- average reward
- control policy
- model free
- policy evaluation
- partially observable markov decision processes
- state action
- state dependent
- state and action spaces
- policy gradient
- model free reinforcement learning
- real time
- agent learns
- continuous state
- machine learning
- temporal difference
- control problems
- learning algorithm
- transition model
- continuous state spaces
- reinforcement learning methods
- infinite horizon
- optimal control
- decision problems
- supervised learning
- partially observable domains
- finite state
- learning problems
- transfer learning
- access control
- temporal difference learning
- supply chain
- learning process
- multi agent
- average cost
- asymptotically optimal