FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classification-猫眼课题宝

DOI：

10.48550/arxiv.2207.10888

发表时间：

2022-07

期刊：

ArXiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Xiao-Ze Lin;Seungbae Kim;Jungseock Joo

现有的修剪技术保留了深度神经网络做出正确预测的整体能力，但也可能在压缩过程中放大隐藏的偏差。我们提出了一种新颖的剪枝方法，即公平感知梯度剪枝方法（FairGRAPE），该方法可以最大限度地减少剪枝对不同子组的不成比例的影响。我们的方法计算每个模型权重的每组重要性，并选择一个权重子集，以在剪枝中保持相对组间总重要性。然后，所提出的方法修剪具有较小重要性值的网络边缘，并通过更新重要性值来重复该过程。我们在四个不同的数据集（FairFace、UTKFace、CelebA 和 ImageNet）上证明了我们的方法对于面部属性分类任务的有效性，其中与最先进的剪枝算法相比，我们的方法将性能下降的差异减少了高达 90%。我们的方法在修剪率较高 (99%) 的环境中显着更有效。实验中使用的代码和数据集可在 https://github.com/Bernardo1998/FairGRAPE 获取

Existing pruning techniques preserve deep neural networks' overall ability to make correct predictions but may also amplify hidden biases during the compression process. We propose a novel pruning method, Fairness-aware GRAdient Pruning mEthod (FairGRAPE), that minimizes the disproportionate impacts of pruning on different sub-groups. Our method calculates the per-group importance of each model weight and selects a subset of weights that maintain the relative between-group total importance in pruning. The proposed method then prunes network edges with small importance values and repeats the procedure by updating importance values. We demonstrate the effectiveness of our method on four different datasets, FairFace, UTKFace, CelebA, and ImageNet, for the tasks of face attribute classification where our method reduces the disparity in performance degradation by up to 90% compared to the state-of-the-art pruning algorithms. Our method is substantially more effective in a setting with a high pruning rate (99%). The code and dataset used in the experiments are available at https://github.com/Bernardo1998/FairGRAPE

FairGRAPE: Fairness-aware GRAdient Pruning mEthod for Face Attribute Classification

文献摘要

相似文献