Weakly-Supervised Reinforcement Learning for Controllable Behavior

Lee Lisa,Eysenbach Benjamin,Salakhutdinov Ruslan大牛学者, Shixiang, Gu,Finn Chelsea大牛学者

NIPS 2020(2020)

引用 20|浏览367
暂无评分
摘要
Reinforcement learning (RL) is a powerful framework for learning to take actions to solve tasks. However, in many settings, an agent must winnow down the inconceivably large space of all possible tasks to the single task that it is currently being asked to solve. Can we instead constrain the space of tasks to those that are semantically meaningful? In this work, we introduce a framework for using weak supervision to automatically disentangle this semantically meaningful subspace of tasks from the enormous space of nonsensical "chaff" tasks. We show that this learned subspace enables efficient exploration and provides a representation that captures distance between states. On a variety of challenging, vision-based continuous control problems, our approach leads to substantial performance gains, particularly as the complexity of the environment grows.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络