Frequency Matters: Explaining Biases of Face Recognition in the Frequency Domain
This addresses bias in face recognition systems, which is a critical issue for fairness in AI applications, but the approach is incremental as it builds on existing frequency-based explanation methods.
The paper tackled the problem of demographic bias in face recognition models by explaining it using frequency-based analysis, finding that different frequencies are important depending on ethnicity.
Face recognition (FR) models are vulnerable to performance variations across demographic groups. The causes for these performance differences are unclear due to the highly complex deep learning-based structure of face recognition models. Several works aimed at exploring possible roots of gender and ethnicity bias, identifying semantic reasons such as hairstyle, make-up, or facial hair as possible sources. Motivated by recent discoveries of the importance of frequency patterns in convolutional neural networks, we explain bias in face recognition using state-of-the-art frequency-based explanations. Our extensive results show that different frequencies are important to FR models depending on the ethnicity of the samples.