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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

The imitation learning of self-driving vehicle policies through behavioralcloning is often carried out in an open-loop fashion, ignoring the effect ofactions to future states. Training such policies purely with Empirical RiskMinimization (ERM) can be detrimental to real-world performance, as it biasespolicy networks towards matching only open-loop behavior, showing poor resultswhen evaluated in closed-loop. In this work, we develop an efficient andsimple-to-implement principle called Closed-loop Weighted Empirical RiskMinimization (CW-ERM), in which a closed-loop evaluation procedure is firstused to identify training data samples that are important for practical drivingperformance and then we these samples to help debias the policy network. Weevaluate CW-ERM in a challenging urban driving dataset and show that thisprocedure yields a significant reduction in collisions as well as othernon-differentiable closed-loop metrics.

CW-ERM: Improving Autonomous Driving Planning with Closed-loop Weighted Empirical Risk Minimization