Mitigating Bias Using Model-Agnostic Data Attribution

TL;DR

This paper introduces a model-agnostic, region-based attribution method using CNNs to identify bias-related image regions, and employs targeted noise to reduce bias in machine learning models, improving fairness.

Contribution

It presents a novel pixel attribution technique for bias mitigation that is model-agnostic and leverages region-specific information to regularize datasets.

Findings

Effective bias reduction in heavily biased datasets

Improved fairness of classifiers trained with targeted noise

Region-based attributions reveal bias-related image areas

Abstract

Mitigating bias in machine learning models is a critical endeavor for ensuring fairness and equity. In this paper, we propose a novel approach to address bias by leveraging pixel image attributions to identify and regularize regions of images containing significant information about bias attributes. Our method utilizes a model-agnostic approach to extract pixel attributions by employing a convolutional neural network (CNN) classifier trained on small image patches. By training the classifier to predict a property of the entire image using only a single patch, we achieve region-based attributions that provide insights into the distribution of important information across the image. We propose utilizing these attributions to introduce targeted noise into datasets with confounding attributes that bias the data, thereby constraining neural networks from learning these biases and emphasizing the primary attributes. Our approach demonstrates its efficacy in enabling the training of unbiased classifiers on heavily biased datasets.