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

Most practical recommender systems focus on estimating immediate userengagement without considering the long-term effects of recommendations on userbehavior. Reinforcement learning (RL) methods offer the potential to optimizerecommendations for long-term user engagement. However, since users are oftenpresented with slates of multiple items - which may have interacting effects onuser choice - methods are required to deal with the combinatorics of the RLaction space. In this work, we address the challenge of making slate-basedrecommendations to optimize long-term value using RL. Our contributions arethree-fold. (i) We develop SLATEQ, a decomposition of value-basedtemporal-difference and Q-learning that renders RL tractable with slates. Undermild assumptions on user choice behavior, we show that the long-term value(LTV) of a slate can be decomposed into a tractable function of its componentitem-wise LTVs. (ii) We outline a methodology that leverages existing myopiclearning-based recommenders to quickly develop a recommender that handles LTV.(iii) We demonstrate our methods in simulation, and validate the scalability ofdecomposed TD-learning using SLATEQ in live experiments on YouTube.

Reinforcement Learning for Slate-based Recommender Systems: A Tractable Decomposition and Practical Methodology