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

Learning from demonstration is widely used as an efficient way for robots toacquire new skills. However, it typically requires that demonstrations providefull access to the state and action sequences. In contrast, learning fromobservation offers a way to utilize unlabeled demonstrations (e.g., video) toperform imitation learning. One approach to this is behavioral cloning fromobservation (BCO). The original implementation of BCO proceeds by firstlearning an inverse dynamics model and then using that model to estimate actionlabels, thereby reducing the problem to behavioral cloning. However, existingapproaches to BCO require a large number of initial interactions in the firststep. Here, we provide a novel theoretical analysis of BCO, introduce amodification BCO*, and show that in the semi-supervised setting, BCO* canconcurrently improve both its estimate for the inverse dynamics model and theexpert policy. This result allows us to eliminate the dependence on initialinteractions and dramatically improve the sample complexity of BCO. We evaluatethe effectiveness of our algorithm through experiments on various benchmarkdomains. The results demonstrate that concurrent training not only improvesover the performance of BCO but also results in performance that is competitivewith state-of-the-art imitation learning methods such as GAIL and Value-Dice.

Concurrent Training Improves the Performance of Behavioral Cloning from Observation