“Coarse-To-Fine Q-Attention: Efficient Learning for Visual Robotic Manipulation via Discretisation”, Stephen James, Kentaro Wada, Tristan Laidlow, Andrew J. Davison2021-06-23 (neural net, robotics; similar)⁠:

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 voxelized scene, coarse-to-fine Q-attention learns what part of the scene to ‘zoom’ into. When this ‘zooming’ behavior is applied iteratively, it results in a near-lossless discretization 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.

Similar Links: