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

We consider contextual bandits with linear constraints (CBwLC), a variant ofcontextual bandits in which the algorithm consumes multiple resources subjectto linear constraints on total consumption. This problem generalizes contextualbandits with knapsacks (CBwK), allowing for packing and covering constraints,as well as positive and negative resource consumption. We provide the firstalgorithm for CBwLC (or CBwK) that is based on regression oracles. Thealgorithm is simple, computationally efficient, and admits vanishing regret. Itis statistically optimal for the variant of CBwK in which the algorithm muststop once some constraint is violated. Further, we provide the firstvanishing-regret guarantees for CBwLC (or CBwK) that extend beyond thestochastic environment. We side-step strong impossibility results from priorwork by identifying a weaker (and, arguably, fairer) benchmark to compareagainst. Our algorithm builds on LagrangeBwK (Immorlica et al., FOCS 2019), aLagrangian-based technique for CBwK, and SquareCB (Foster and Rakhlin, ICML2020), a regression-based technique for contextual bandits. Our analysisleverages the inherent modularity of both techniques.

Contextual Bandits with Packing and Covering Constraints: A Modular Lagrangian Approach via Regression