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

The $K$-armed dueling bandits problem, where the feedback is in the form ofnoisy pairwise preferences, has been widely studied due its applications ininformation retrieval, recommendation systems, etc. Motivated by concerns thatuser preferences/tastes can evolve over time, we consider the problem ofdueling bandits with distribution shifts. Specifically, we study the recentnotion of significant shifts (Suk and Kpotufe, 2022), and ask whether one candesign an adaptive algorithm for the dueling problem with$O(\sqrt{K\tilde{L}T})$ dynamic regret, where $\tilde{L}$ is the (unknown)number of significant shifts in preferences. We show that the answer to thisquestion depends on the properties of underlying preference distributions.  Firstly, we give an impossibility result that rules out any algorithm with$O(\sqrt{K\tilde{L}T})$ dynamic regret under the well-studied Condorcet and SSTclasses of preference distributions. Secondly, we show that $\text{SST} \cap\text{STI}$ is the largest amongst popular classes of preference distributionswhere it is possible to design such an algorithm. Overall, our results providesan almost complete resolution of the above question for the hierarchy ofdistribution classes.

When Can We Track Significant Preference Shifts in Dueling Bandits?