Features, logits, and labels are the three primary data when a sample passes through a deep neural network. Feature perturbation and label perturbation receive increasing attention in recent years. They have been proven to be useful in various deep learning approaches. For example, (adversarial) feature perturbation can improve the robustness or even generalization capability of learned models. However, limited studies have explicitly explored for the perturbation of logit vectors. This work discusses several existing methods related to class-level logit perturbation. A unified viewpoint between positive/negative data augmentation and loss variations incurred by logit perturbation is established. A theoretical analysis is provided to illuminate why class-level logit perturbation is useful. Accordingly, new methodologies are proposed to explicitly learn to perturb logits for both single-label and multi-label classification tasks. Extensive experiments on benchmark image classification data sets and their long-tail versions indicated the competitive performance of our learning method. As it only perturbs on logit, it can be used as a plug-in to fuse with any existing classification algorithms. All the codes are available at https://github.com/limengyang1992/lpl.
翻译:当样本通过深神经网络时,其特征、登录和标签是三种初级数据:当样本通过深神经网络时,其特性、特征、特征、特征和标签的扰动作用近年来受到越来越多的关注。这些特征、特征、特征、特征、特征和标签的扰动已被证明在各种深层学习方法中有用。例如,(对抗性)特征的扰动可以提高所学模型的稳健性,甚至概括性能力。然而,对于对对登录矢量的扰动进行了明确的探索。这项工作讨论了与分类层次对向矢量的渗透有关的几种现有方法。在正/负性数据增强和对登录扰动引起的损失变化之间建立了统一的观点。提供了理论分析,以说明为何等级对分类有用。因此,提出了新的方法,以明确学习对单标签和多标签分类任务进行扰动性记录。关于基准图像分类数据集及其长性版本的广泛实验表明我们学习方法的竞争性表现。由于对日志仅进行perturb,因此它可以用作任何现有分类/complus/eng AMs。