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

Online learning algorithms, widely used to power search and contentoptimization on the web, must balance exploration and exploitation, potentiallysacrificing the experience of current users for information that will lead tobetter decisions in the future. Recently, concerns have been raised aboutwhether the process of exploration could be viewed as unfair, placing too muchburden on certain individuals or groups. Motivated by these concerns, weinitiate the study of the externalities of exploration - the undesirable sideeffects that the presence of one party may impose on another - under the linearcontextual bandits model. We introduce the notion of a group externality,measuring the extent to which the presence of one population of users impactsthe rewards of another. We show that this impact can in some cases be negative,and that, in a certain sense, no algorithm can avoid it. We then studyexternalities at the individual level, interpreting the act of exploration asan externality imposed on the current user of a system by future users. Thisdrives us to ask under what conditions inherent diversity in the data makesexplicit exploration unnecessary. We build on a recent line of work on thesmoothed analysis of the greedy algorithm that always chooses the action thatcurrently looks optimal, improving on prior results to show that a greedyapproach almost matches the best possible Bayesian regret rate of any otheralgorithm on the same problem instance whenever the diversity conditions hold,and that this regret is at most $\tilde{O}(T^{1/3})$. Returning to group-leveleffects, we show that under the same conditions, negative group externalitiesessentially vanish under the greedy algorithm. Together, our results uncover asharp contrast between the high externalities that exist in the worst case, andthe ability to remove all externalities if the data is sufficiently diverse.

The Externalities of Exploration and How Data Diversity Helps Exploitation