Domain Adversarial Reinforcement Learning

02/14/2021
by   Bonnie Li, et al.
8

We consider the problem of generalization in reinforcement learning where visual aspects of the observations might differ, e.g. when there are different backgrounds or change in contrast, brightness, etc. We assume that our agent has access to only a few of the MDPs from the MDP distribution during training. The performance of the agent is then reported on new unknown test domains drawn from the distribution (e.g. unseen backgrounds). For this "zero-shot RL" task, we enforce invariance of the learned representations to visual domains via a domain adversarial optimization process. We empirically show that this approach allows achieving a significant generalization improvement to new unseen domains.

READ FULL TEXT

page 5

page 7

research
08/17/2020

Zero Shot Domain Generalization

Standard supervised learning setting assumes that training data and test...
research
06/15/2017

Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning

As a step towards developing zero-shot task generalization capabilities ...
research
06/05/2023

Explore to Generalize in Zero-Shot RL

We study zero-shot generalization in reinforcement learning - optimizing...
research
04/07/2021

Unsupervised Visual Attention and Invariance for Reinforcement Learning

Vision-based reinforcement learning (RL) is successful, but how to gener...
research
06/20/2020

Representation via Representations: Domain Generalization via Adversarially Learned Invariant Representations

We investigate the power of censoring techniques, first developed for le...
research
01/27/2020

Rotation, Translation, and Cropping for Zero-Shot Generalization

Deep Reinforcement Learning (DRL) has shown impressive performance on do...
research
07/19/2018

Multitask Reinforcement Learning for Zero-shot Generalization with Subtask Dependencies

We introduce a new RL problem where the agent is required to execute a g...

Please sign up or login with your details

Forgot password? Click here to reset