Skip to yearly menu bar Skip to main content


Spotlight Poster

Distribution-Free Statistical Dispersion Control for Societal Applications

Zhun Deng · Thomas Zollo · Jake Snell · Toniann Pitassi · Richard Zemel

Great Hall & Hall B1+B2 (level 1) #1907

Abstract:

Explicit finite-sample statistical guarantees on model performance are an important ingredient in responsible machine learning. Previous work has focused mainly on bounding either the expected loss of a predictor or the probability that an individual prediction will incur a loss value in a specified range. However, for many high-stakes applications it is crucial to understand and control the \textit{dispersion} of a loss distribution, or the extent to which different members of a population experience unequal effects of algorithmic decisions. We initiate the study of distribution-free control of statistical dispersion measures with societal implications and propose a simple yet flexible framework that allows us to handle a much richer class of statistical functionals beyond previous work. Our methods are verified through experiments in toxic comment detection, medical imaging, and film recommendation.

Chat is not available.