Investigating the Relationship Between Debiasing and Artifact Removal using Saliency Maps
This addresses fairness concerns in AI by revealing connections between debiasing and artifact removal, though it appears incremental as it repurposes existing techniques.
The paper investigates the relationship between debiasing and artifact removal in neural networks for computer vision, showing that successful debiasing redirects model focus away from protected attributes and that artifact removal techniques can improve fairness.
The widespread adoption of machine learning systems has raised critical concerns about fairness and bias, making mitigating harmful biases essential for AI development. In this paper, we investigate the relationship between debiasing and removing artifacts in neural networks for computer vision tasks. First, we introduce a set of novel XAI-based metrics that analyze saliency maps to assess shifts in a model's decision-making process. Then, we demonstrate that successful debiasing methods systematically redirect model focus away from protected attributes. Finally, we show that techniques originally developed for artifact removal can be effectively repurposed for improving fairness. These findings provide evidence for the existence of a bidirectional connection between ensuring fairness and removing artifacts corresponding to protected attributes.