Measuring Diversity

Search results that reflect historic inequities can amplify stereotypes and perpetuate under-representation. Carefully measuring diversity in data sets can help.

Measuring Diversity Explorables Measuring Diversity Search results that reflect historic inequities can amplify stereotypes and perpetuate under-representation. Carefully measuring diversity in data sets can help. Search, ranking and recommendation systems can help find useful documents in large datasets. However, these datasets reflect the biases of the society in which they were created and the systems risk re-entrenching those biases. For example, if someone who is not a white man searches for “CEO pictures” and sees a page of white men , they may feel that only white men can be CEOs, furth

related reading

• Ensuring Diversity: Strategies to Avoid Stereotypes in Business Modelslinkedin.com
• Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrievalarxiv.org
• How well do truth probes generalise? — LessWronglesswrong.com
• Reconciling the accuracy-diversity trade-off in recommendationsarxiv.org
• Measuring Fairnesspair.withgoogle.com
• Biasjzhao.xyz
• Scaling Laws Do Not Scalearxiv.org
• Datasets Have Worldviewspair.withgoogle.com
• Relative notions of fairnessfairmlbook.org
• Statistical Fairness - Turing Commonsalan-turing-institute.github.io
• Biases: An Introductionreadthesequences.com
• [arxiv] an algorithmic framework for fairness elicitationarxiv.org