Link Search Menu Expand Document

CLOUD: Contrastive Learning of Unsupervised Dynamics

Paper PDF Code

Authors

Jianren Wang (Carnegie Mellon University)*; Yujie Lu (Tencent); Hang Zhao (Waymo)

Interactive Session

2020-11-18, 11:50 - 12:20 PST | PheedLoop Session

Abstract

Developing agents that can perform complex control tasks from high dimensional observations such as pixels is challenging due to difficulties in learning dynamics efficiently. In this work, we propose to learn forward and inverse dynamics in a fully unsupervised manner via contrastive estimation. Specifically, we train a forward dynamics model and an inverse dynamics model in the feature space of states and actions with data collected from random exploration. Unlike most existing deterministic models, our energy-based model takes into account the stochastic nature of agent-environment interactions. We demonstrate the efficacy of our approach across a variety of tasks including goal-directed planning and imitation from observations.

Video

Reviews and Rebuttal

Reviews & Rebuttal


Conference on Robot Learning 2020