Mining bias-target Alignment from Voronoi Cells. (arXiv:2305.03691v1 [cs.LG])

Despite significant research efforts, deep neural networks are still
vulnerable to biases: this raises concerns about their fairness and limits
their generalization. In this paper, we propose a bias-agnostic approach to
mitigate the impact of bias in deep neural networks. Unlike traditional
debiasing approaches, we rely on a metric to quantify ``bias
alignment/misalignment'' on target classes, and use this information to
discourage the propagation of bias-target alignment information through the
network. We conduct experiments on several commonly used datasets for …

alignment arxiv bias biases cells fairness impact mining networks neural networks paper raises research voronoi vulnerable