Run Peng
Run Peng
Home
Publications
CV
Light
Dark
Automatic
Reinforcement Learning
CommonGrid: Building Common Ground through Belief Maintenance in Situated Communication
Constructing a multi-agent, collaboration benchmark which requires belief and intention modeling; Training RL agents to perform theory of mind modeling during the collaboration`
Learning Exploration Policies with View-based Intrinsic Rewards
Intrinsic Reward Design, Reinforcement Learning, Sample Efficiency in Exploration
Yijie Guo
,
Yao Fu
,
Run Peng
,
Honglak Lee
PDF
Cite
Cite
×