View full-text article in PMC PeerJ Comput Sci. 2024 Sep 30;10:e2317. doi: 10.7717/peerj-cs.2317 Search in PMC Search in PubMed View in NLM Catalog Add to search Copyright and License information © 2024 Strotherm et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited. PMC Copyright notice Table 2. Overview of empirical fairness definitions. Comparison of our generalized empirical fairness definitions (definition 2.7, 2.8, 2.14 and 2.15) and the corresponding existing definitions (e.g., Pessach & Shmueli, 2022). Derived generalized empirical definitions (multi cases) Existing empirical definitions (binary cases) Independence: Y={0,1}, S arbitrary: Y=S={0,1}: DI minsk1,sk2∈Sℙ(Y^=1|S=sk1)ℙ(Y^=1|S=sk2) ℙ(Y^=1|S=0)ℙ(Y^=1|S=1) Independence: Y,S arbitrary: Y=S={0,1}: DP maxy∈Y,sk1,sk2∈S |ℙ(Y^=y|S=sk1)−ℙ(Y^=y|S=sk2)| |ℙ(Y^=1|S=0)−ℙ(Y^=1|S=1)| Separation: Y={0,1}, S arbitrary: Y=S={0,1}: EO maxsk1,sk2∈S |ℙ(Y^=1|S=sk1,Y=1)−ℙ(Y^=1|S=sk2,Y=1)| |ℙ(Y^=1|S=0,Y=1)−ℙ(Y^=1|S=1,Y=1)| Separation: Y,S arbitrary, ∀y∈Y: Y=S={0,1}: EOs maxsk1,sk2∈S |ℙ(Y^=y|S=sk1,Y=y)−ℙ(Y^=y|S=sk2,Y=y)| |ℙ(Y^=1|S=0,Y=1)−ℙ(Y^=1|S=1,Y=1)| |ℙ(Y^=1|S=0,Y=0)−ℙ(Y^=1|S=1,Y=0)|