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

In recent years, machine learning algorithms have become ubiquitous in amultitude of high-stakes decision-making applications. The unparalleled abilityof machine learning algorithms to learn patterns from data also enables them toincorporate biases embedded within. A biased model can then make decisions thatdisproportionately harm certain groups in society -- limiting their access tofinancial services, for example. The awareness of this problem has given riseto the field of Fair ML, which focuses on studying, measuring, and mitigatingunfairness in algorithmic prediction, with respect to a set of protected groups(e.g., race or gender). However, the underlying causes for algorithmicunfairness still remain elusive, with researchers divided between blamingeither the ML algorithms or the data they are trained on. In this work, wemaintain that algorithmic unfairness stems from interactions between models andbiases in the data, rather than from isolated contributions of either of them.To this end, we propose a taxonomy to characterize data bias and we study a setof hypotheses regarding the fairness-accuracy trade-offs that fairness-blind MLalgorithms exhibit under different data bias settings. On our real-worldaccount-opening fraud use case, we find that each setting entails specifictrade-offs, affecting fairness in expected value and variance -- the latteroften going unnoticed. Moreover, we show how algorithms compare differently interms of accuracy and fairness, depending on the biases affecting the data.Finally, we note that under specific data bias conditions, simplepre-processing interventions can successfully balance group-wise error rates,while the same techniques fail in more complex settings.

Understanding Unfairness in Fraud Detection through Model and Data Bias Interactions