Shedding light on underrepresentation and Sampling Bias in machine learning
This work addresses fairness assessment in ML by highlighting how bias affects different groups unequally, challenging common mitigation strategies, but it is incremental in refining existing concepts.
The paper tackles the problem of accurately measuring discrimination in machine learning models by clarifying sampling bias into sample size bias and underrepresentation bias, and shows that discrimination can be decomposed into variance, bias, and noise.
Accurately measuring discrimination is crucial to faithfully assessing fairness of trained machine learning (ML) models. Any bias in measuring discrimination leads to either amplification or underestimation of the existing disparity. Several sources of bias exist and it is assumed that bias resulting from machine learning is born equally by different groups (e.g. females vs males, whites vs blacks, etc.). If, however, bias is born differently by different groups, it may exacerbate discrimination against specific sub-populations. Sampling bias, is inconsistently used in the literature to describe bias due to the sampling procedure. In this paper, we attempt to disambiguate this term by introducing clearly defined variants of sampling bias, namely, sample size bias (SSB) and underrepresentation bias (URB). We show also how discrimination can be decomposed into variance, bias, and noise. Finally, we challenge the commonly accepted mitigation approach that discrimination can be addressed by collecting more samples of the underrepresented group.