Deep Semi-Supervised Anomaly Detection

Abstract

Deep approaches to anomaly detection have recently shown promising results over shallow methods on large and complex datasets. Typically anomaly detection is treated as an unsupervised learning problem. In practice however, one may have---in addition to a large set of unlabeled samples---access to a small pool of labeled samples, e.g. a subset verified by some domain expert as being normal or anomalous. Semi-supervised approaches to anomaly detection aim to utilize such labeled samples, but most proposed methods are limited to merely including labeled normal samples. Only a few methods take advantage of labeled anomalies, with existing deep approaches being domain-specific. In this work we present Deep SAD, an end-to-end deep methodology for general semi-supervised anomaly detection. We further introduce an information-theoretic framework for deep anomaly detection based on the idea that the entropy of the latent distribution for normal data should be lower than the entropy of the anomalous distribution, which can serve as a theoretical interpretation for our method. In extensive experiments on MNIST, Fashion-MNIST, and CIFAR-10, along with other anomaly detection benchmark datasets, we demonstrate that our method is on par or outperforms shallow, hybrid, and deep competitors, yielding appreciable performance improvements even when provided with only little labeled data.

arXiv comment23 pages, Published as a conference paper at ICLR 2020

References in corpus (16)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Auto-Encoding Variational Bayes
• The information bottleneck method
• Learning deep representations by mutual information estimation and maximization
• Semi-Supervised Learning with Deep Generative Models
• Deep Learning for Anomaly Detection: A Survey
• Opening the Black Box of Deep Neural Networks via Information
• Semi-Supervised Learning with Ladder Networks
• Semi-Supervised Learning with Generative Adversarial Networks
• Deep Anomaly Detection with Outlier Exposure
• Deep Anomaly Detection Using Geometric Transformations
• Unsupervised and Semi-supervised Anomaly Detection with LSTM Neural Networks
• Toward Supervised Anomaly Detection
• InfoVAE: Information Maximizing Variational Autoencoders
• Information based clustering
• Anomaly Detection using One-Class Neural Networks

Cited by in corpus (11)

• A Comprehensive Survey on Graph Anomaly Detection with Deep Learning
• Artificial Intelligence based Anomaly Detection of Energy Consumption in Buildings: A Review, Current Trends and New Perspectives
• Quasi Anomalous Knowledge: Searching for new physics with embedded knowledge
• Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Features
• RLAD: Time Series Anomaly Detection through Reinforcement Learning and Active Learning
• Neural Transformation Learning for Deep Anomaly Detection Beyond Images
• Novelty Detection Via Blurring
• TimeAutoML: Autonomous Representation Learning for Multivariate Irregularly Sampled Time Series
• Meta-SVDD: Probabilistic Meta-Learning for One-Class Classification in Cancer Histology Images
• Anomaly Detection and Automated Labeling for Voter Registration File Changes
• Anomaly-Injected Deep Support Vector Data Description for Text Outlier Detection