Timezone: »

Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning
Junhyuk Oh · Satinder Singh · Honglak Lee · Pushmeet Kohli

Tue Aug 08 08:48 PM -- 09:06 PM (PDT) @ C4.5

As a step towards developing zero-shot task generalization capabilities in reinforcement learning (RL), we introduce a new RL problem where the agent should learn to execute sequences of instructions after learning useful skills that solve subtasks. In this problem, we consider two types of generalizations: to previously unseen instructions and to longer sequences of instructions. For generalization over unseen instructions, we propose a new objective which encourages learning correspondences between similar subtasks by making analogies. For generalization over sequential instructions, we present a hierarchical architecture where a meta controller learns to use the acquired skills for executing the instructions. To deal with delayed reward, we propose a new neural architecture in the meta controller that learns when to update the subtask, which makes learning more efficient. Experimental results on a stochastic 3D domain show that the proposed ideas are crucial for generalization to longer instructions as well as unseen instructions.

Author Information

Junhyuk Oh (University of Michigan)
Satinder Singh (University of Michigan)
Honglak Lee (Google / U. Michigan)
Pushmeet Kohli (Microsoft Research)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors