Sparsity-based Safety Conservatism for Constrained Offline Reinforcement Learning.
Minjae ChoChuangchuang SunPublished in: CoRR (2024)
Keyphrases
- reinforcement learning
- function approximation
- real time
- state space
- sparse representation
- learning algorithm
- dynamic programming
- data sets
- reinforcement learning algorithms
- high dimensional
- temporal difference
- optimal policy
- sparse approximation
- markov decision processes
- linear combination
- supervised learning
- multi agent
- transfer learning
- safety critical
- multi agent reinforcement learning
- objective function