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

While training fair machine learning models has been studied extensively inrecent years, most developed methods rely on the assumption that the trainingand test data have similar distributions. In the presence of distributionshifts, fair models may behave unfairly on test data. There have been somedevelopments for fair learning robust to distribution shifts to address thisshortcoming. However, most proposed solutions are based on the assumption ofhaving access to the causal graph describing the interaction of differentfeatures. Moreover, existing algorithms require full access to data and cannotbe used when small batches are used (stochastic/batch implementation). Thispaper proposes the first stochastic distributionally robust fairness frameworkwith convergence guarantees that do not require knowledge of the causal graph.More specifically, we formulate the fair inference in the presence of thedistribution shift as a distributionally robust optimization problem under$L_p$ norm uncertainty sets with respect to the Exponential Renyi MutualInformation (ERMI) as the measure of fairness violation. We then discuss howthe proposed method can be implemented in a stochastic fashion. We haveevaluated the presented framework's performance and efficiency throughextensive experiments on real datasets consisting of distribution shifts.

Dr. FERMI: A Stochastic Distributionally Robust Fair Empirical Risk Minimization Framework