D4RL: Datasets for Deep Data-Driven Reinforcement Learning

TL;DR

This paper introduces new offline RL benchmarks with diverse datasets to better evaluate and advance offline reinforcement learning algorithms, addressing limitations of previous benchmarks.

Contribution

The authors present a set of offline RL benchmarks with varied datasets, along with evaluation protocols and open-source tools to facilitate research and identify algorithm shortcomings.

Findings

Existing algorithms reveal significant deficiencies on new benchmarks.

Diverse datasets expose limitations of current offline RL methods.

Benchmark resources are publicly available for community use.

Abstract

The offline reinforcement learning (RL) setting (also known as full batch RL), where a policy is learned from a static dataset, is compelling as progress enables RL methods to take advantage of large, previously-collected datasets, much like how the rise of large datasets has fueled results in supervised learning. However, existing online RL benchmarks are not tailored towards the offline setting and existing offline RL benchmarks are restricted to data generated by partially-trained agents, making progress in offline RL difficult to measure. In this work, we introduce benchmarks specifically designed for the offline setting, guided by key properties of datasets relevant to real-world applications of offline RL. With a focus on dataset collection, examples of such properties include: datasets generated via hand-designed controllers and human demonstrators, multitask datasets where an agent performs different tasks in the same environment, and datasets collected with mixtures of policies. By moving beyond simple benchmark tasks and data collected by partially-trained RL agents, we reveal important and unappreciated deficiencies of existing algorithms. To facilitate research, we have released our benchmark tasks and datasets with a comprehensive evaluation of existing algorithms, an evaluation protocol, and open-source examples. This serves as a common starting point for the community to identify shortcomings in existing offline RL methods and a collaborative route for progress in this emerging area.