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

Practical Imitation Learning (IL) systems rely on large human demonstrationdatasets for successful policy learning. However, challenges lie in maintainingthe quality of collected data and addressing the suboptimal nature of somedemonstrations, which can compromise the overall dataset quality and hence thelearning outcome. Furthermore, the intrinsic heterogeneity in human behaviorcan produce equally successful but disparate demonstrations, furtherexacerbating the challenge of discerning demonstration quality. To addressthese challenges, this paper introduces Learning to Discern (L2D), an offlineimitation learning framework for learning from demonstrations with diversequality and style. Given a small batch of demonstrations with sparse qualitylabels, we learn a latent representation for temporally embedded trajectorysegments. Preference learning in this latent space trains a quality evaluatorthat generalizes to new demonstrators exhibiting different styles. Empirically,we show that L2D can effectively assess and learn from varying demonstrations,thereby leading to improved policy performance across a range of tasks in bothsimulations and on a physical robot.

Learning to Discern: Imitating Heterogeneous Human Demonstrations with Preference and Representation Learning