Augmentation Curriculum Learning For Generalization in RL

Dylan Yung, Andrew Szot, Prithvijit Chattopadhyay, Judy Hoffman, Zsolt Kira

Keywords: reinforcement learning, generalization, pixel-based RL, embodied learning

TL;DR: Combining data augmentation, reinforcement learning and curriculum learning for generalization in reinforcement learning

Abstract:

Many Reinforcement Learning tasks rely solely on pixel-based observations of
the environment. During deployment, these observations can fall victim to visual
perturbations and distortions, causing the agent’s policy to significantly degrade
in performance. This motivates the need for robust agents that can generalize in
the face of visual distribution shift. One common technique for doing this is to ap-
ply augmentations during training; however, it comes at the cost of performance.
We propose Augmentation Curriculum Learning a novel curriculum learning ap-
proach that schedules augmentation into training into a weak augmentation phase
and strong augmentation phase. We also introduce a novel visual augmentation
strategy that proves to aid in the benchmarks we evaluate on. Our method achieves
state-of-the-art performance on Deep Mind Control Generalization Benchmark.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Reinforcement Learning (eg, decision and control, planning, hierarchical RL, robotics)