利用von Mises-Fisher混合模型缓解人脸识别中的性别偏见
Mitigating Gender Bias in Face Recognition Using the von Mises-Fisher Mixture Model
AI总结：
本文针对深度人脸识别中的性别偏见问题，提出基于Fair von Mises-Fisher损失的后处理方法，通过变换预训练嵌入增强受歧视子群体的表示能力，并引入BFAR和BFRR指标验证其显著降低性别偏见。
摘要：
尽管深度学习算法在广泛的日常应用中表现出高性能和可靠性，但许多研究表明，大量模型存在偏见，会歧视人群中的特定子群体（如性别、族裔）。这促使从业者开发公平系统，使其在敏感群体上具有一致或可比的性能。本文研究了深度人脸识别网络的性别偏见。为了衡量这种偏见，我们引入了两个新指标BFAR和BFRR，它们更好地反映了人脸识别系统固有的部署需求。受几何考虑的启发，我们通过一种新的后处理方法缓解性别偏见，该方法对预训练模型的深度嵌入进行变换，为受歧视的子群体提供更强的表示能力。该方法通过最小化Fair von Mises-Fisher损失来训练一个浅层神经网络，其超参数考虑了每个性别的类内方差。有趣的是，我们通过实验观察到这些超参数与我们的公平性指标相关。事实上，在多种数据集上进行的大量数值实验表明，精心选择超参数能显著减少性别偏见。实验所用代码可在https://github.com/JRConti/EthicalModule_vMF找到。
英文摘要：
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.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
