Publication

Self-Supervised Policy Adaptation during Deployment

Conference Article

Conference

International Conference on Learning Representations (ICLR)

Edition

Ninth

Pages

1-18

Doc link

https://openreview.net/forum?id=o_V-MjyyGV_

File

Download the digital copy of the doc pdf document

Authors

Projects associated

Abstract

In most real world scenarios, a policy trained by reinforcement learning in one environment needs to be deployed in another, potentially quite different environment. However, generalization across different environments is known to be hard. A natural solution would be to keep training after deployment in the new environment, but this cannot be done if the new environment offers no reward signal. Our work explores the use of self-supervision to allow the policy to continue training after deployment without using any rewards. While previous methods explicitly anticipate changes in the new environment, we assume no prior knowledge of those changes yet still obtain significant improvements. Empirical evaluations are performed on diverse simulation environments from DeepMind Control suite and ViZDoom, as well as real robotic manipulation tasks in continuously changing environments, taking observations from an uncalibrated camera. Our method improves generalization in 31 out of 36 environments across various tasks and outperforms domain randomization on a majority of environments. Webpage and implementation: https://nicklashansen.github.io/PAD/.

Categories

artificial intelligence.

Scientific reference

N.H. , R. Jangir, Y.S. , G. Alenyà, P. Abbeel, A.A. , L.P. and X.W.. Self-Supervised Policy Adaptation during Deployment, Ninth International Conference on Learning Representations, 2021, Online, pp. 1-18.