Diversity-Driven Exploration Strategy for Deep Reinforcement Learning.
Zhang-Wei HongTzu-Yun ShannShih-Yang SuYi-Hsiang ChangTsu-Jui FuChun-Yi LeePublished in: NeurIPS (2018)
Keyphrases
- exploration strategy
- reinforcement learning
- unknown environments
- function approximation
- state space
- dynamic programming
- markov decision processes
- locally optimal
- learning algorithm
- markov decision problems
- reinforcement learning algorithms
- optimal policy
- multi agent
- temporal difference
- machine learning
- single agent
- free space
- optimal control