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

A generalist robot equipped with learned skills must be able to perform manytasks in many different environments. However, zero-shot generalization to newsettings is not always possible. When the robot encounters a new environment orobject, it may need to finetune some of its previously learned skills toaccommodate this change. But crucially, previously learned behaviors and modelsshould still be suitable to accelerate this relearning. In this paper, we aimto study how generative models of possible outcomes can allow a robot to learnvisual representations of affordances, so that the robot can sample potentiallypossible outcomes in new situations, and then further train its policy toachieve those outcomes. In effect, prior data is used to learn what kinds ofoutcomes may be possible, such that when the robot encounters an unfamiliarsetting, it can sample potential outcomes from its model, attempt to reachthem, and thereby update both its skills and its outcome model. This approach,visuomotor affordance learning (VAL), can be used to train goal-conditionedpolicies that operate on raw image inputs, and can rapidly learn to manipulatenew objects via our proposed affordance-directed exploration scheme. We showthat VAL can utilize prior data to solve real-world tasks such drawer opening,grasping, and placing objects in new scenes with only five minutes of onlineexperience in the new scene.

What Can I Do Here? Learning New Skills by Imagining Visual Affordances