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

In recent years, there has been a growing interest in utilizing reinforcementlearning (RL) to optimize long-term rewards in recommender systems. Sinceindustrial recommender systems are typically designed as multi-stage systems,RL methods with a single agent face challenges when optimizing multiple stagessimultaneously. The reason is that different stages have different observationspaces, and thus cannot be modeled by a single agent. To address this issue, wepropose a novel UNidirectional-EXecution-based multi-agent ReinforcementLearning (UNEX-RL) framework to reinforce the long-term rewards in multi-stagerecommender systems. We show that the unidirectional execution is a key featureof multi-stage recommender systems, bringing new challenges to the applicationsof multi-agent reinforcement learning (MARL), namely the observation dependencyand the cascading effect. To tackle these challenges, we provide a cascadinginformation chain (CIC) method to separate the independent observations fromaction-dependent observations and use CIC to train UNEX-RL effectively. We alsodiscuss practical variance reduction techniques for UNEX-RL. Finally, we showthe effectiveness of UNEX-RL on both public datasets and an online recommendersystem with over 100 million users. Specifically, UNEX-RL reveals a 0.558%increase in users' usage time compared with single-agent RL algorithms inonline A/B experiments, highlighting the effectiveness of UNEX-RL in industrialrecommender systems.

UNEX-RL: Reinforcing Long-Term Rewards in Multi-Stage Recommender Systems with UNidirectional EXecution