KoGuN: Accelerating Deep Reinforcement Learning via Integrating Human Suboptimal Knowledge.
Peng ZhangJianye HaoWeixun WangHongyao TangYi MaYihai DuanYan ZhengPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- domain knowledge
- human beings
- human experts
- prior knowledge
- human users
- function approximation
- learning process
- human knowledge
- human cognition
- knowledge based systems
- learning systems
- cognitive map
- computational models
- background knowledge
- knowledge management
- state space
- dynamic programming
- multi agent