# Tilted Cross Entropy (TCE): Promoting Fairness in Semantic Segmentation

Source: [Tilted Cross Entropy (TCE): Promoting Fairness in Semantic Segmentation](<https://www.sota2.com/research/paper/68fa699d1e209131cea23152>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T23:04:46.235Z

Authors: Attila Szabo, Hadi Jamali-Rad, Siva-Datta Mannava

Published: 2021-03-25T18:00:50.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2103.14051>)

[PDF](<https://arxiv.org/pdf/2103.14051.pdf>)