Offline Reinforcement Learning as Anti-exploration.
Shideh RezaeifarRobert DadashiNino VieillardLéonard HussenotOlivier BachemOlivier PietquinMatthieu GeistPublished in: AAAI (2022)
Keyphrases
- reinforcement learning
- active exploration
- exploration strategy
- action selection
- model based reinforcement learning
- function approximation
- exploration exploitation
- state space
- policy search
- autonomous learning
- model free
- real time
- learning algorithm
- multi agent
- learning process
- exploration exploitation tradeoff
- markov decision processes
- neural network
- reinforcement learning algorithms
- temporal difference learning
- machine learning
- balancing exploration and exploitation
- optimal control
- temporal difference
- information visualization
- transfer learning
- supervised learning
- dynamic programming
- search algorithm
- security protection