Improving Fairness using Vision-Language Driven Image Augmentation

TL;DR

This paper introduces a novel image augmentation method using a diffusion model guided by vision-language prompts to mitigate biases related to age and skin color in facial datasets, thereby enhancing fairness in deep learning models.

Contribution

It proposes a new approach to improve fairness by editing protected attributes in images through semantic paths in a diffusion model, guided by contrastive text supervision.

Findings

Augmented images reduce bias in model predictions.

Method improves overall accuracy and fairness metrics.

Effective on CelebA-HQ and UTKFace datasets.

Abstract

Fairness is crucial when training a deep-learning discriminative model, especially in the facial domain. Models tend to correlate specific characteristics (such as age and skin color) with unrelated attributes (downstream tasks), resulting in biases which do not correspond to reality. It is common knowledge that these correlations are present in the data and are then transferred to the models during training. This paper proposes a method to mitigate these correlations to improve fairness. To do so, we learn interpretable and meaningful paths lying in the semantic space of a pre-trained diffusion model (DiffAE) -- such paths being supervised by contrastive text dipoles. That is, we learn to edit protected characteristics (age and skin color). These paths are then applied to augment images to improve the fairness of a given dataset. We test the proposed method on CelebA-HQ and UTKFace on several downstream tasks with age and skin color as protected characteristics. As a proxy for fairness, we compute the difference in accuracy with respect to the protected characteristics. Quantitative results show how the augmented images help the model improve the overall accuracy, the aforementioned metric, and the disparity of equal opportunity. Code is available at: https://github.com/Moreno98/Vision-Language-Bias-Control.