Learning State-Specific Action Masks for Reinforcement Learning.
Ziyi WangXinran LiLuoyang SunHaifeng ZhangHualin LiuJun WangPublished in: Algorithms (2024)
Keyphrases
- reinforcement learning
- learning algorithm
- state action
- learning problems
- supervised learning
- action sequences
- action selection
- state space
- knowledge acquisition
- agent learns
- domain specific
- machine learning
- hidden state
- reinforcement learning methods
- reward signal
- online learning
- state abstraction
- evolutionary learning
- bayesian networks
- partially observable
- reinforcement learning algorithms
- function approximation
- learning tasks
- markov decision processes
- learning systems