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

When the performance of a machine learning model varies over groups definedby sensitive attributes (e.g., gender or ethnicity), the performance disparitycan be expressed in terms of the probability distributions of the input andoutput variables over each group. In this paper, we exploit this fact to reducethe disparate impact of a fixed classification model over a population ofinterest. Given a black-box classifier, we aim to eliminate the performance gapby perturbing the distribution of input variables for the disadvantaged group.We refer to the perturbed distribution as a counterfactual distribution, andcharacterize its properties for common fairness criteria. We introduce adescent algorithm to learn a counterfactual distribution from data. We thendiscuss how the estimated distribution can be used to build a data preprocessorthat can reduce disparate impact without training a new model. We validate ourapproach through experiments on real-world datasets, showing that it can repairdifferent forms of disparity without a significant drop in accuracy.

Repairing without Retraining: Avoiding Disparate Impact with Counterfactual Distributions