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

We propose a novel master-slave architecture to solve the top-$K$combinatorial multi-armed bandits problem with non-linear bandit feedback anddiversity constraints, which, to the best of our knowledge, is the firstcombinatorial bandits setting considering diversity constraints under banditfeedback. Specifically, to efficiently explore the combinatorial andconstrained action space, we introduce six slave models with distinguishedmerits to generate diversified samples well balancing rewards and constraintsas well as efficiency. Moreover, we propose teacher learning based optimizationand the policy co-training technique to boost the performance of the multipleslave models. The master model then collects the elite samples provided by theslave models and selects the best sample estimated by a neural contextualUCB-based network to make a decision with a trade-off between exploration andexploitation. Thanks to the elaborate design of slave models, the co-trainingmechanism among slave models, and the novel interactions between the master andslave models, our approach significantly surpasses existing state-of-the-artalgorithms in both synthetic and real datasets for recommendation tasks. Thecode is available at:\url{https://github.com/huanghanchi/Master-slave-Algorithm-for-Top-K-Bandits}.

Master-slave Deep Architecture for Top-K Multi-armed Bandits with Non-linear Bandit Feedback and Diversity Constraints