Incorporating Explanations to Balance the Exploration and Exploitation of Deep Reinforcement Learning.
Xinzhi WangYang LiuYudong ChangChao JiangQingjie ZhangPublished in: KSEM (2) (2022)
Keyphrases
- exploration exploitation tradeoff
- reinforcement learning
- active exploration
- function approximation
- exploration strategy
- objective function
- relevance feedback
- action selection
- state space
- exploration exploitation
- active learning
- model based reinforcement learning
- search capabilities
- learning algorithm
- stochastic approximation
- temporal difference
- reinforcement learning algorithms
- learning capabilities
- model free
- markov decision processes
- generating explanations
- dynamic programming
- real time
- hidden markov models
- temporal difference learning
- reinforcement learning methods
- multi agent reinforcement learning
- decision making
- multi agent
- search engine
- robotic control