Keywords: Offline Reinforcement Learning, Data Collection, Reachability, Unsupervised Learning, Curiosity-Driven Learning
TL;DR: We propose a novel adaptive reachability-based method to improve the data collection process in offline reinforcement learning.
Abstract: In offline reinforcement learning (RL), while the majority of efforts are focusing on engineering sophisticated learning algorithms given a fixed dataset, very few works have been carried out to improve the dataset quality itself. More importantly, it is even challenging to collect a task-agnostic dataset such that the offline RL agent can learn multiple skills from it. In this paper, we propose a Curiosity-driven Unsupervised Data Collection (CUDC) method to improve the data collection process. Specifically, we quantify the agent's internal belief to estimate the probability of the k-step future states being reachable from the current states. Different from existing approaches that implicitly assume limited feature space with fixed environment steps, CUDC is capable of adapting the number of environment steps to explore. Thus, the feature representation can be substantially diversified with the dynamics information. With this adaptive reachability mechanism in place, the agent can navigate itself to collect higher-quality data with curiosity. Empirically, CUDC surpasses existing unsupervised methods in sample efficiency and learning performance in various downstream offline RL tasks of the DeepMind control suite.
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)