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/

Zendy Plus

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

Zendy Tools

Zendy Open

In this paper, we improve the regret bound for online kernel selection underbandit feedback. Previous algorithm enjoys a $O((\Vertf\Vert^2_{\mathcal{H}_i}+1)K^{\frac{1}{3}}T^{\frac{2}{3}})$ expected bound forLipschitz loss functions. We prove two types of regret bounds improving theprevious bound. For smooth loss functions, we propose an algorithm with a$O(U^{\frac{2}{3}}K^{-\frac{1}{3}}(\sum^K_{i=1}L_T(f^\ast_i))^{\frac{2}{3}})$expected bound where $L_T(f^\ast_i)$ is the cumulative losses of optimalhypothesis in $\mathbb{H}_{i}=\{f\in\mathcal{H}_i:\Vertf\Vert_{\mathcal{H}_i}\leq U\}$. The data-dependent bound keeps the previousworst-case bound and is smaller if most of candidate kernels match well withthe data. For Lipschitz loss functions, we propose an algorithm with a$O(U\sqrt{KT}\ln^{\frac{2}{3}}{T})$ expected bound asymptotically improving theprevious bound. We apply the two algorithms to online kernel selection withtime constraint and prove new regret bounds matching or improving the previous$O(\sqrt{T\ln{K}} +\Vertf\Vert^2_{\mathcal{H}_i}\max\{\sqrt{T},\frac{T}{\sqrt{\mathcal{R}}}\})$expected bound where $\mathcal{R}$ is the time budget. Finally, we empiricallyverify our algorithms on online regression and classification tasks.

Improved Regret Bounds for Online Kernel Selection under Bandit Feedback