Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation

Abstract

Anomaly detection suffers from unbalanced data since anomalies are quite rare. Synthetically generated anomalies are a solution to such ill or not fully defined data. However, synthesis requires an expressive representation to guarantee the quality of the generated data. In this paper, we propose a two-level hierarchical latent space representation that distills inliers' feature-descriptors (through autoencoders) into more robust representations based on a variational family of distributions (through a variational autoencoder) for zero-shot anomaly generation. From the learned latent distributions, we select those that lie on the outskirts of the training data as synthetic-outlier generators. And, we synthesize from them, i.e., generate negative samples without seen them before, to train binary classifiers. We found that the use of the proposed hierarchical structure for feature distillation and fusion creates robust and general representations that allow us to synthesize pseudo outlier samples. And in turn, train robust binary classifiers for true outlier detection (without the need for actual outliers during training). We demonstrate the performance of our proposal on several benchmarks for anomaly detection.

To appear in IEEE Trans. on Neural Networks and Learning Systems

References in corpus (16)

• Auto-Encoding Variational Bayes
• The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions
• Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications
• Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Images
• A geometric analysis of subspace clustering with outliers
• Deep Anomaly Detection Using Geometric Transformations
• Deep Structured Energy Based Models for Anomaly Detection
• Robust PCA via Outlier Pursuit
• Generative Probabilistic Novelty Detection with Adversarial Autoencoders
• Coherence Pursuit: Fast, Simple, and Robust Principal Component Analysis
• Robust computation of linear models by convex relaxation
• Learning Deep Representations of Appearance and Motion for Anomalous Event Detection
• Anomaly Detection for Skin Disease Images Using Variational Autoencoder
• Video Anomaly Detection and Localization via Gaussian Mixture Fully Convolutional Variational Autoencoder
• Connectivity-Optimized Representation Learning via Persistent Homology
• Latent Laplacian Maximum Entropy Discrimination for Detection of High-Utility Anomalies

Cited by in corpus (4)

• Improving Generalizability of Graph Anomaly Detection Models via Data Augmentation
• Generative Adversarial Network-based Cross-Project Fault Prediction
• Adversarial Stacked Auto-Encoders for Fair Representation Learning
• What augmentations are sensitive to hyper-parameters and why?