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

Imitation learning has proven to be useful for many real-world problems, butapproaches such as behavioral cloning suffer from data mismatch and compoundingerror issues. One attempt to address these limitations is the DAgger algorithm,which uses the state distribution induced by the novice to sample correctiveactions from the expert. Such sampling schemes, however, require the expert toprovide action labels without being fully in control of the system. This candecrease safety and, when using humans as experts, is likely to degrade thequality of the collected labels due to perceived actuator lag. In this work, wepropose HG-DAgger, a variant of DAgger that is more suitable for interactiveimitation learning from human experts in real-world systems. In addition totraining a novice policy, HG-DAgger also learns a safety threshold for amodel-uncertainty-based risk metric that can be used to predict the performanceof the fully trained novice in different regions of the state space. Weevaluate our method on both a simulated and real-world autonomous driving task,and demonstrate improved performance over both DAgger and behavioral cloning.

HG-DAgger: Interactive Imitation Learning with Human Experts