BAdd: Bias Mitigation through Bias Addition

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sarridis, Ioannis, Koutlis, Christos, Papadopoulos, Symeon, Diou, Christos
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916363957698560
author Sarridis, Ioannis
Koutlis, Christos
Papadopoulos, Symeon
Diou, Christos
author_facet Sarridis, Ioannis
Koutlis, Christos
Papadopoulos, Symeon
Diou, Christos
contents Computer vision (CV) datasets often exhibit biases that are perpetuated by deep learning models. While recent efforts aim to mitigate these biases and foster fair representations, they fail in complex real-world scenarios. In particular, existing methods excel in controlled experiments involving benchmarks with single-attribute injected biases, but struggle with multi-attribute biases being present in well-established CV datasets. Here, we introduce BAdd, a simple yet effective method that allows for learning fair representations invariant to the attributes introducing bias by incorporating features representing these attributes into the backbone. BAdd is evaluated on seven benchmarks and exhibits competitive performance, surpassing state-of-the-art methods on both single- and multi-attribute benchmarks. Notably, BAdd achieves +27.5% and +5.5% absolute accuracy improvements on the challenging multi-attribute benchmarks, FB-Biased-MNIST and CelebA, respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2408_11439
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle BAdd: Bias Mitigation through Bias Addition
Sarridis, Ioannis
Koutlis, Christos
Papadopoulos, Symeon
Diou, Christos
Computer Vision and Pattern Recognition
Computer vision (CV) datasets often exhibit biases that are perpetuated by deep learning models. While recent efforts aim to mitigate these biases and foster fair representations, they fail in complex real-world scenarios. In particular, existing methods excel in controlled experiments involving benchmarks with single-attribute injected biases, but struggle with multi-attribute biases being present in well-established CV datasets. Here, we introduce BAdd, a simple yet effective method that allows for learning fair representations invariant to the attributes introducing bias by incorporating features representing these attributes into the backbone. BAdd is evaluated on seven benchmarks and exhibits competitive performance, surpassing state-of-the-art methods on both single- and multi-attribute benchmarks. Notably, BAdd achieves +27.5% and +5.5% absolute accuracy improvements on the challenging multi-attribute benchmarks, FB-Biased-MNIST and CelebA, respectively.
title BAdd: Bias Mitigation through Bias Addition
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.11439