Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper addresses the challenge of achieving fairness in classification without relying on demographic information by proposing a novel minimax-fair method called SPECTRE. SPECTRE adjusts the spectrum of a Fourier feature mapping and constrains the deviation of the worst-case distribution from the empirical distribution, mitigating the over-pessimism of existing robust optimization techniques. Empirical results on American Community Survey datasets across 20 states demonstrate that SPECTRE achieves superior fairness guarantees and robustness compared to state-of-the-art methods, even those with access to demographic data.
Achieve state-of-the-art fairness guarantees in classification *without* demographic data using SPECTRE, a novel spectral adjustment method that outperforms even demographically-aware approaches.
As automated classification systems become increasingly prevalent, concerns have emerged over their potential to reinforce and amplify existing societal biases. In the light of this issue, many methods have been proposed to enhance the fairness guarantees of classifiers. Most of the existing interventions assume access to group information for all instances, a requirement rarely met in practice. Fairness without access to demographic information has often been approached through robust optimization techniques,which target worst-case outcomes over a set of plausible distributions known as the uncertainty set. However, their effectiveness is strongly influenced by the chosen uncertainty set. In fact, existing approaches often overemphasize outliers or overly pessimistic scenarios, compromising both overall performance and fairness. To overcome these limitations, we introduce SPECTRE, a minimax-fair method that adjusts the spectrum of a simple Fourier feature mapping and constrains the extent to which the worst-case distribution can deviate from the empirical distribution. We perform extensive experiments on the American Community Survey datasets involving 20 states. The safeness of SPECTRE comes as it provides the highest average values on fairness guarantees together with the smallest interquartile range in comparison to state-of-the-art approaches, even compared to those with access to demographic group information. In addition, we provide a theoretical analysis that derives computable bounds on the worst-case error for both individual groups and the overall population, as well as characterizes the worst-case distributions responsible for these extremal performances