Understanding and Addressing the Pitfalls of Bisimulation-based Representations in Offline Reinforcement Learning.
Hongyu ZangXin LiLeiji ZhangYang LiuBaigui SunRiashat IslamRemi Tachet des CombesRomain LarochePublished in: NeurIPS (2023)
Keyphrases
- reinforcement learning
- external representations
- state space
- learning algorithm
- dynamic programming
- function approximation
- real time
- machine learning
- website
- neural network
- search algorithm
- higher level
- cognitive processing
- multi agent
- database
- optimal policy
- markov decision processes
- representation scheme
- function approximators
- temporal difference learning