论文标题
DROCC：深度强大的一级分类
DROCC: Deep Robust One-Class Classification
论文作者
论文摘要
一级问题（例如一级SVM和隔离森林）的经典方法需要仔细的特征工程，当应用于图像等结构化域时。最先进的方法旨在通过两种主要方法来利用深度学习来学习适当的功能。基于预测转换的第一种方法（Golan＆El-Yaniv，2018; Hendrycks等，2019a）虽然在某些领域成功，但至关重要地取决于一般很难获得的适当域特异性转换集。最小化经典的一级损失的第二种方法，例如，DeepSVDD（Ruff等，2018）遭受了表示崩溃的基本缺点。在这项工作中，我们提出了深度鲁棒的一级分类（DROCC），既适用于大多数标准域，而无需任何侧面信息和强大的表示即可崩溃。 DROCC是基于以下假设：兴趣类别的点位于一个局部采样，局部线性的低维歧管上。经验评估表明，DROCC在两个不同的单级问题设置以及跨不同领域的一系列现实数据集中非常有效：表格数据，图像（CIFAR和Imagenet），音频和时间序列，可在整体检测中提供高达20％的准确性。代码可在https://github.com/microsoft/edgeml上找到。
Classical approaches for one-class problems such as one-class SVM and isolation forest require careful feature engineering when applied to structured domains like images. State-of-the-art methods aim to leverage deep learning to learn appropriate features via two main approaches. The first approach based on predicting transformations (Golan & El-Yaniv, 2018; Hendrycks et al., 2019a) while successful in some domains, crucially depends on an appropriate domain-specific set of transformations that are hard to obtain in general. The second approach of minimizing a classical one-class loss on the learned final layer representations, e.g., DeepSVDD (Ruff et al., 2018) suffers from the fundamental drawback of representation collapse. In this work, we propose Deep Robust One-Class Classification (DROCC) that is both applicable to most standard domains without requiring any side-information and robust to representation collapse. DROCC is based on the assumption that the points from the class of interest lie on a well-sampled, locally linear low dimensional manifold. Empirical evaluation demonstrates that DROCC is highly effective in two different one-class problem settings and on a range of real-world datasets across different domains: tabular data, images (CIFAR and ImageNet), audio, and time-series, offering up to 20% increase in accuracy over the state-of-the-art in anomaly detection. Code is available at https://github.com/microsoft/EdgeML.
