Reinforcement Learning via Auxillary Task Distillation
Abhinav N Harish*, Larry Heck, Josiah P Hanna, Zsolt Kira, Andrew Szot
Abstract
"We present Reinforcement Learning via Auxiliary Task Distillation (AuxDistill), a new method that enables reinforcement learning (RL) to perform long-horizon robot control problems by distilling behaviors from auxiliary RL tasks. AuxDistill achieves this by concurrently carrying out multi-task RL with auxiliary tasks, which are easier to learn and relevant to the main task. A weighted distillation loss transfers behaviors from these auxiliary tasks to solve the main task. We demonstrate that AuxDistill can learn a pixels-to-actions policy for a challenging multi-stage embodied object rearrangement task from the environment reward without demonstrations, a learning curriculum, or pre-trained skills. AuxDistill achieves 2.3× higher success than the previous state-of-the-art baseline in the Habitat Object Rearrangement benchmark and outperforms methods that use pre-trained skills and expert demonstrations."
BibTeX
@inproceedings{eccv2024_reinforcementlea,
title = {Reinforcement Learning via Auxillary Task Distillation},
author = {Abhinav N Harish* and Larry Heck and Josiah P Hanna and Zsolt Kira and Andrew Szot},
booktitle = {ECCV 2024},
year = {2024}
}