End-to-End Convolutional Activation Anomaly Analysis for Anomaly Detection

TL;DR

This paper introduces E2E-CA$^3$, an end-to-end convolutional autoencoder-based anomaly detection method that effectively detects anomalies in image and tabular data with a lightweight architecture.

Contribution

It extends the A$^3$ anomaly detection approach by integrating convolutional autoencoders and combining classification and reconstruction losses for broader applicability.

Findings

Effective on datasets like MNIST, CIFAR-10, KDDcup99

Lightweight yet promising anomaly detection performance

Versatile application to image and tabular data

Abstract

We propose an End-to-end Convolutional Activation Anomaly Analysis (E2E-CA$^3$), which is a significant extension of A$^3$ anomaly detection approach proposed by Sperl, Schulze and B\"ottinger, both in terms of architecture and scope of application. In contrast to the original idea, we utilize a convolutional autoencoder as a target network, which allows for natural application of the method both to image and tabular data. The alarm network is also designed as a CNN, where the activations of convolutional layers from CAE are stacked together into $k+1-$dimensional tensor. Moreover, we combine the classification loss of the alarm network with the reconstruction error of the target CAE, as a "best of both worlds" approach, which greatly increases the versatility of the network. The evaluation shows that despite generally straightforward and lightweight architecture, it has a very promising anomaly detection performance on common datasets such as MNIST, CIFAR-10 and KDDcup99.