Learn to Effectively Explore in Context-Based Meta-RL

Zhang Jin
Zhang Jin
Wang Jianhao
Wang Jianhao
Hu Hao
Hu Hao
Zhang Chongjie
Zhang Chongjie
Cited by: 0|Bibtex|Views39|Links

Abstract:

Meta reinforcement learning (meta-RL) provides a principled approach for fast adaptation to novel tasks by extracting prior knowledge from previous tasks. Under such settings, it is crucial for the agent to perform efficient exploration during adaptation to collect useful experiences. However, existing methods suffer from poor adaptatio...More

Code:

Data:

Your rating :
0

 

Tags
Comments