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

One of the main issues in Imitation Learning is the erroneous behavior of anagent when facing out-of-distribution situations, not covered by the set ofdemonstrations given by the expert. In this work, we tackle this problem byintroducing a novel active learning and control algorithm, SAFARI. Duringtraining, it allows an agent to request further human demonstrations when theseout-of-distribution situations are met. At deployment, it combines model-freeacting using behavioural cloning with model-based planning to reducestate-distribution shift, using future state reconstruction as a test for statefamiliarity. We empirically demonstrate how this method increases theperformance on a set of manipulation tasks with respect to passive ImitationLearning, by gathering more informative demonstrations and by minimizingstate-distribution shift at test time. We also show how this method enables theagent to autonomously predict failure rapidly and safely.

SAFARI: Safe and Active Robot Imitation Learning with Imagination