Coarse-to-Fine Q-attention: Efficient Learning for Visual Robotic Manipulation via Discretisation

TL;DR

This paper introduces a coarse-to-fine Q-attention method that discretizes the scene for efficient reinforcement learning in robotic manipulation, achieving state-of-the-art results with minimal data and rapid training.

Contribution

It presents a novel discretization approach enabling discrete RL in robotics, improving stability and data efficiency over actor-critic methods.

Findings

Achieves state-of-the-art performance on RLBench tasks.

Trains real-world policies in minutes with few demonstrations.

Enables near-lossless discretization of translation space.

Abstract

We present a coarse-to-fine discretisation method that enables the use of discrete reinforcement learning approaches in place of unstable and data-inefficient actor-critic methods in continuous robotics domains. This approach builds on the recently released ARM algorithm, which replaces the continuous next-best pose agent with a discrete one, with coarse-to-fine Q-attention. Given a voxelised scene, coarse-to-fine Q-attention learns what part of the scene to 'zoom' into. When this 'zooming' behaviour is applied iteratively, it results in a near-lossless discretisation of the translation space, and allows the use of a discrete action, deep Q-learning method. We show that our new coarse-to-fine algorithm achieves state-of-the-art performance on several difficult sparsely rewarded RLBench vision-based robotics tasks, and can train real-world policies, tabula rasa, in a matter of minutes, with as little as 3 demonstrations.