English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present a coarse-to-fine discretisation method that enables the use ofdiscrete reinforcement learning approaches in place of unstable anddata-inefficient actor-critic methods in continuous robotics domains. Thisapproach builds on the recently released ARM algorithm, which replaces thecontinuous next-best pose agent with a discrete one, with coarse-to-fineQ-attention. Given a voxelised scene, coarse-to-fine Q-attention learns whatpart of the scene to 'zoom' into. When this 'zooming' behaviour is appliediteratively, it results in a near-lossless discretisation of the translationspace, and allows the use of a discrete action, deep Q-learning method. We showthat our new coarse-to-fine algorithm achieves state-of-the-art performance onseveral difficult sparsely rewarded RLBench vision-based robotics tasks, andcan train real-world policies, tabula rasa, in a matter of minutes, with aslittle as 3 demonstrations.

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