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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We consider the problem of learning multi-stage vision-based tasks on a realrobot from a single video of a human performing the task, while leveragingdemonstration data of subtasks with other objects. This problem presents anumber of major challenges. Video demonstrations without teleoperation are easyfor humans to provide, but do not provide any direct supervision. Learningpolicies from raw pixels enables full generality but calls for large functionapproximators with many parameters to be learned. Finally, compound tasks canrequire impractical amounts of demonstration data, when treated as a monolithicskill. To address these challenges, we propose a method that learns both how tolearn primitive behaviors from video demonstrations and how to dynamicallycompose these behaviors to perform multi-stage tasks by "watching" a humandemonstrator. Our results on a simulated Sawyer robot and real PR2 robotillustrate our method for learning a variety of order fulfillment and kitchenserving tasks with novel objects and raw pixel inputs.

One-Shot Hierarchical Imitation Learning of Compound Visuomotor Tasks