A Minimalist Approach to Offline Reinforcement Learning.
Scott FujimotoShixiang Shane GuPublished in: NeurIPS (2021)
Keyphrases
- reinforcement learning
- function approximation
- state space
- model free
- reinforcement learning algorithms
- multi agent
- optimal policy
- dynamic programming
- learning process
- real time
- learning algorithm
- multi agent systems
- supervised learning
- markov decision processes
- policy search
- multi agent reinforcement learning
- transition model
- evolutionary learning
- direct policy search
- temporal difference learning
- markov decision process
- robot control
- partially observable
- temporal difference
- objective function
- information systems
- computer vision
- real world