): near-optimal safety-constrained reinforcement learning in polynomial time.
David M. BossensNicholas BishopPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- special case
- deterministic domains
- learning algorithm
- function approximation
- state space
- worst case
- approximation algorithms
- temporal difference learning
- optimal policy
- supervised learning
- model free
- computational complexity
- reinforcement learning methods
- machine learning
- real time
- optimal control
- action selection
- reinforcement learning algorithms
- stochastic approximation
- multi agent reinforcement learning