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

We investigate the problem of reliably assessing group fairness when labeledexamples are few but unlabeled examples are plentiful. We propose a generalBayesian framework that can augment labeled data with unlabeled data to producemore accurate and lower-variance estimates compared to methods based on labeleddata alone. Our approach estimates calibrated scores for unlabeled examples ineach group using a hierarchical latent variable model conditioned on labeledexamples. This in turn allows for inference of posterior distributions withassociated notions of uncertainty for a variety of group fairness metrics. Wedemonstrate that our approach leads to significant and consistent reductions inestimation error across multiple well-known fairness datasets, sensitiveattributes, and predictive models. The results show the benefits of using bothunlabeled data and Bayesian inference in terms of assessing whether aprediction model is fair or not.

Can I Trust My Fairness Metric? Assessing Fairness with Unlabeled Data and Bayesian Inference