Neural networks achieve the state-of-the-art in image classification tasks. However, they can encode spurious variations or biases that may be present in the training data. For example, training an age predictor on a dataset that is not balanced for gender can lead to gender biased predicitons (e.g. wrongly predicting that males are older if only elderly males are in the training set). We present two distinct contributions: 1) An algorithm that can remove multiple sources of variation from the feature representation of a network. We demonstrate that this algorithm can be used to remove biases from the feature representation, and thereby improve classification accuracies, when training networks on extremely biased datasets. 2) An ancestral origin database of 14,000 images of individuals from East Asia, the Indian subcontinent, sub-Saharan Africa, and Western Europe. We demonstrate on this dataset, for a number of facial attribute classification tasks, that we are able to remove racial biases from the network feature representation.
翻译:神经网络在图像分类任务中达到了最先进的艺术水平。 但是,它们可以将培训数据中可能存在的虚假变异或偏差进行分类。例如,在性别不平衡的数据集上培训年龄预测员,可能导致性别偏见的先决论(例如错误地预测,如果只有老年男性参加培训,男性的年龄就会老化)。我们提出两种不同的贡献:1) 一种算法,可以消除网络特征代表的多种变异来源。我们证明,在培训网络使用极端偏差数据集时,这种算法可以用来消除特征代表的偏差,从而改进分类的精度。2 一个由来自东亚、印度次大陆、撒哈拉以南非洲和西欧的14 000个人图像组成的祖传来源数据库。我们在这个数据集上展示了面部属性分类的一些任务,即我们能够从网络特征代表中消除种族偏见。