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

Importance sampling (IS) is often used to perform off-policy policyevaluation but is prone to several issues, especially when the behavior policyis unknown and must be estimated from data. Significant differences between thetarget and behavior policies can result in uncertain value estimates due to,for example, high variance and non-evaluated actions. If the behavior policy isestimated using black-box models, it can be hard to diagnose potential problemsand to determine for which inputs the policies differ in their suggestedactions and resulting values. To address this, we propose estimating thebehavior policy for IS using prototype learning. We apply this approach in theevaluation of policies for sepsis treatment, demonstrating how the prototypesgive a condensed summary of differences between the target and behaviorpolicies while retaining an accuracy comparable to baseline estimators. We alsodescribe estimated values in terms of the prototypes to better understand whichparts of the target policies have the most impact on the estimates. Using asimulator, we study the bias resulting from restricting models to useprototypes.

Case-based off-policy policy evaluation using prototype learning