CVDec 12, 2018

Considering Race a Problem of Transfer Learning

arXiv:1812.04751v19 citations
Originality Incremental advance
AI Analysis

This addresses fairness and accuracy issues in biometric systems for diverse populations, but it is incremental as it applies existing transfer learning concepts to race-specific domains.

The paper tackles performance disparities in biometric applications across racial subgroups by framing race as a transfer learning boundary for facial classification and image synthesis, demonstrating improved classification transfer and showing that GANs trained on one race perform poorly on another using a new race annotation for CelebA.

As biometric applications are fielded to serve large population groups, issues of performance differences between individual sub-groups are becoming increasingly important. In this paper we examine cases where we believe race is one such factor. We look in particular at two forms of problem; facial classification and image synthesis. We take the novel approach of considering race as a boundary for transfer learning in both the task (facial classification) and the domain (synthesis over distinct datasets). We demonstrate a series of techniques to improve transfer learning of facial classification; outperforming similar models trained in the target's own domain. We conduct a study to evaluate the performance drop of Generative Adversarial Networks trained to conduct image synthesis, in this process, we produce a new annotation for the Celeb-A dataset by race. These networks are trained solely on one race and tested on another - demonstrating the subsets of the CelebA to be distinct domains for this task.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes