Anomaly Detection with Robust Deep Autoencoders

TLDR

Deep autoencoders excel at extracting nonlinear features, yet their performance degrades in the presence of large outliers and pervasive noise when clean training data is unavailable. The authors propose extensions to deep autoencoders that simultaneously learn high‑quality features and remove outliers and noise without requiring any clean data. Their Robust Deep Autoencoder (RDA) model, inspired by Robust PCA, decomposes input data into a low‑rank component reconstructed by a deep autoencoder and a sparse component capturing outliers, while a Group Robust variant (GRDA) incorporates grouped sparsity norms to separate random anomalies from structured corruptions. Experiments on benchmark datasets demonstrate that GRDA achieves superior anomaly detection performance compared to conventional methods.

Abstract

Deep autoencoders, and other deep neural networks, have demonstrated their effectiveness in discovering non-linear features across many problem domains. However, in many real-world problems, large outliers and pervasive noise are commonplace, and one may not have access to clean training data as required by standard deep denoising autoencoders. Herein, we demonstrate novel extensions to deep autoencoders which not only maintain a deep autoencoders' ability to discover high quality, non-linear features but can also eliminate outliers and noise without access to any clean training data. Our model is inspired by Robust Principal Component Analysis, and we split the input data X into two parts, $X = L_{D} + S$, where $L_{D}$ can be effectively reconstructed by a deep autoencoder and $S$ contains the outliers and noise in the original data X. Since such splitting increases the robustness of standard deep autoencoders, we name our model a "Robust Deep Autoencoder (RDA)". Further, we present generalizations of our results to grouped sparsity norms which allow one to distinguish random anomalies from other types of structured corruptions, such as a collection of features being corrupted across many instances or a collection of instances having more corruptions than their fellows. Such "Group Robust Deep Autoencoders (GRDA)" give rise to novel anomaly detection approaches whose superior performance we demonstrate on a selection of benchmark problems.