Saved in:
Bibliographic Details
Main Authors: Minch, Allen, Vu, Hung Anh, Warren, Anne Marie
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.02012
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917558217605120
author Minch, Allen
Vu, Hung Anh
Warren, Anne Marie
author_facet Minch, Allen
Vu, Hung Anh
Warren, Anne Marie
contents This project explores adversarial training techniques to develop fairer Deep Neural Networks (DNNs) to mitigate the inherent bias they are known to exhibit. DNNs are susceptible to inheriting bias with respect to sensitive attributes such as race and gender, which can lead to life-altering outcomes (e.g., demographic bias in facial recognition software used to arrest a suspect). We propose a robust optimization problem, which we demonstrate can improve fairness in several datasets, both synthetic and real-world, using an affine linear model. Leveraging second order information, we are able to find a solution to our optimization problem more efficiently than a purely first order method.
format Preprint
id arxiv_https___arxiv_org_abs_2401_02012
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Fast & Fair: Efficient Second-Order Robust Optimization for Fairness in Machine Learning
Minch, Allen
Vu, Hung Anh
Warren, Anne Marie
Machine Learning
Computers and Society
Numerical Analysis
65F10, 65F22, 65K05, 90C47
This project explores adversarial training techniques to develop fairer Deep Neural Networks (DNNs) to mitigate the inherent bias they are known to exhibit. DNNs are susceptible to inheriting bias with respect to sensitive attributes such as race and gender, which can lead to life-altering outcomes (e.g., demographic bias in facial recognition software used to arrest a suspect). We propose a robust optimization problem, which we demonstrate can improve fairness in several datasets, both synthetic and real-world, using an affine linear model. Leveraging second order information, we are able to find a solution to our optimization problem more efficiently than a purely first order method.
title Fast & Fair: Efficient Second-Order Robust Optimization for Fairness in Machine Learning
topic Machine Learning
Computers and Society
Numerical Analysis
65F10, 65F22, 65K05, 90C47
url https://arxiv.org/abs/2401.02012