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

It is commonplace to use data containing personal information to buildpredictive models in the framework of empirical risk minimization (ERM). Whilethese models can be highly accurate in prediction, results obtained from thesemodels with the use of sensitive data may be susceptible to privacy attacks.Differential privacy (DP) is an appealing framework for addressing such dataprivacy issues by providing mathematically provable bounds on the privacy lossincurred when releasing information from sensitive data. Previous work hasprimarily concentrated on applying DP to unweighted ERM. We consider animportant generalization to weighted ERM (wERM). In wERM, each individual'scontribution to the objective function can be assigned varying weights. In thiscontext, we propose the first differentially private wERM algorithm, backed bya rigorous theoretical proof of its DP guarantees under mild regularityconditions. Extending the existing DP-ERM procedures to wERM paves a path toderiving privacy-preserving learning methods for individualized treatmentrules, including the popular outcome weighted learning (OWL). We evaluate theperformance of the DP-wERM application to OWL in a simulation study and in areal clinical trial of melatonin for sleep health. All empirical resultsdemonstrate the viability of training OWL models via wERM with DP guaranteeswhile maintaining sufficiently useful model performance. Therefore, werecommend practitioners consider implementing the proposed privacy-preservingOWL procedure in real-world scenarios involving sensitive data.

A Differentially Private Weighted Empirical Risk Minimization Procedure and its Application to Outcome Weighted Learning