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

Off-Policy reinforcement learning has been a driving force for thestate-of-the-art conversational AIs leading to more natural humanagentinteractions and improving the user satisfaction for goal-oriented agents.However, in large-scale commercial settings, it is often challenging to balancebetween policy improvements and experience continuity on the broad spectrum ofapplications handled by such system. In the literature, off-policy evaluationand guard-railing on aggregate statistics has been commonly used to addressthis problem. In this paper, we propose a method for curating and leveraginghigh-precision samples sourced from historical regression incident reports tovalidate, safe-guard, and improve policies prior to the online deployment. Weconducted extensive experiments using data from a real-world conversationalsystem and actual regression incidents. The proposed method is currentlydeployed in our production system to protect customers against brokenexperiences and enable long-term policy improvements.

Scalable and Safe Remediation of Defective Actions in Self-Learning Conversational Systems