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

Parameter-free algorithms are online learning algorithms that do not requiresetting learning rates. They achieve optimal regret with respect to thedistance between the initial point and any competitor. Yet, parameter-freealgorithms do not take into account the geometry of the losses. Recently, inthe stochastic optimization literature, it has been proposed to instead usetruncated linear lower bounds, which produce better performance by more closelymodeling the losses. In particular, truncated linear models greatly reduce theproblem of overshooting the minimum of the loss function. Unfortunately,truncated linear models cannot be used with parameter-free algorithms becausethe updates become very expensive to compute. In this paper, we propose newparameter-free algorithms that can take advantage of truncated linear modelsthrough a new update that has an "implicit" flavor. Based on a noveldecomposition of the regret, the new update is efficient, requires only onegradient at each step, never overshoots the minimum of the truncated model, andretains the favorable parameter-free properties. We also conduct an empiricalstudy demonstrating the practical utility of our algorithms.

Implicit Parameter-free Online Learning with Truncated Linear Models