# Mitigating Gender Bias in Face Recognition Using the von Mises-Fisher Mixture Model Source: [Mitigating Gender Bias in Face Recognition Using the von Mises-Fisher Mixture Model]() Collection index: [llms.txt]() Data updated: 2026-01-02T04:58:35.845Z Authors: Jean-R\\'emy Conti, Nathan Noiry, Vincent Despiegel, St\\'ephane Gentric, St\\'ephan Cl\\'emen\\c{c}on Published: 2022-10-24T23:53:56.000Z ## Abstract In spite of the high performance and reliability of deep learning algorithms in a wide range of everyday applications, many investigations tend to show that a lot of models exhibit biases, discriminating against specific subgroups of the population (e.g. gender, ethnicity). This urges the practitioner to develop fair systems with a uniform/comparable performance across sensitive groups. In this work, we investigate the gender bias of deep Face Recognition networks. In order to measure this bias, we introduce two new metrics, $\\mathrm{BFAR}$ and $\\mathrm{BFRR}$, that better reflect the inherent deployment needs of Face Recognition systems. Motivated by geometric considerations, we mitigate gender bias through a new post-processing methodology which transforms the deep embeddings of a pre-trained model to give more representation power to discriminated subgroups. It consists in training a shallow neural network by minimizing a Fair von Mises-Fisher loss whose hyperparameters account for the intra-class variance of each gender. Interestingly, we empirically observe that these hyperparameters are correlated with our fairness metrics. In fact, extensive numerical experiments on a variety of datasets show that a careful selection significantly reduces gender bias. The code used for the experiments can be found at https://github.com/JRConti/EthicalModule\_vMF. ## 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. - [Face Recognition on LFW 1.0 (full set)](): Face Recognition; LFW 1.0 (full set); FRR @ FAR: 5.8. - [Face Verification on IJB-C 1:1 protocol (test)](): Face Verification; IJB-C 1:1 protocol (test); FRR @ FAR=1e-4 (%): 0.0434. ## Original sources and implementation [Original paper]() [PDF]() [Implementation]()