Tilted Cross Entropy (TCE): Promoting Fairness in Semantic Segmentation
This addresses fairness issues in semantic segmentation for computer vision applications, but it is incremental as it adapts an existing tilted ERM method to a specific task.
The paper tackled unfair performance disparities among classes in semantic segmentation by proposing a tilted cross-entropy loss, which improved low-performing classes and overall fairness on Cityscapes and ADE20k datasets.
Traditional empirical risk minimization (ERM) for semantic segmentation can disproportionately advantage or disadvantage certain target classes in favor of an (unfair but) improved overall performance. Inspired by the recently introduced tilted ERM (TERM), we propose tilted cross-entropy (TCE) loss and adapt it to the semantic segmentation setting to minimize performance disparity among target classes and promote fairness. Through quantitative and qualitative performance analyses, we demonstrate that the proposed Stochastic TCE for semantic segmentation can efficiently improve the low-performing classes of Cityscapes and ADE20k datasets trained with multi-class cross-entropy (MCCE), and also results in improved overall fairness.