Batch Exploration with Examples for Scalable Robotic Reinforcement Learning

TL;DR

This paper introduces BEE, a scalable exploration method guided by minimal human supervision, which improves data collection and downstream task performance in vision-based robotic manipulation.

Contribution

The paper proposes BEE, a novel exploration technique using human-provided example images to focus exploration on relevant state regions, enhancing data efficiency and task performance.

Findings

BEE explores relevant objects more than twice as often as baseline methods.

BEE improves downstream task performance in vision-based robotic manipulation.

BEE is effective in both simulation and real robot environments.

Abstract

Learning from diverse offline datasets is a promising path towards learning general purpose robotic agents. However, a core challenge in this paradigm lies in collecting large amounts of meaningful data, while not depending on a human in the loop for data collection. One way to address this challenge is through task-agnostic exploration, where an agent attempts to explore without a task-specific reward function, and collect data that can be useful for any downstream task. While these approaches have shown some promise in simple domains, they often struggle to explore the relevant regions of the state space in more challenging settings, such as vision based robotic manipulation. This challenge stems from an objective that encourages exploring everything in a potentially vast state space. To mitigate this challenge, we propose to focus exploration on the important parts of the state space using weak human supervision. Concretely, we propose an exploration technique, Batch Exploration with Examples (BEE), that explores relevant regions of the state-space, guided by a modest number of human provided images of important states. These human provided images only need to be collected once at the beginning of data collection and can be collected in a matter of minutes, allowing us to scalably collect diverse datasets, which can then be combined with any batch RL algorithm. We find that BEE is able to tackle challenging vision-based manipulation tasks both in simulation and on a real Franka robot, and observe that compared to task-agnostic and weakly-supervised exploration techniques, it (1) interacts more than twice as often with relevant objects, and (2) improves downstream task performance when used in conjunction with offline RL.