Improving auto-encoder novelty detection using channel attention and entropy minimization

TL;DR

This paper enhances auto-encoder based novelty detection by integrating channel attention and entropy minimization to better reconstruct normal samples and distinguish outliers.

Contribution

It introduces an attention mechanism and entropy-based constraints into auto-encoders, improving their ability to detect novel or abnormal samples.

Findings

Achieves comparable performance on three public datasets

Improves focus on inlier samples through attention mechanism

Enhances latent space sparsity with entropy minimization

Abstract

Novelty detection is a important research area which mainly solves the classification problem of inliers which usually consists of normal samples and outliers composed of abnormal samples. Auto-encoder is often used for novelty detection. However, the generalization ability of the auto-encoder may cause the undesirable reconstruction of abnormal elements and reduce the identification ability of the model. To solve the problem, we focus on the perspective of better reconstructing the normal samples as well as retaining the unique information of normal samples to improve the performance of auto-encoder for novelty detection. Firstly, we introduce attention mechanism into the task. Under the action of attention mechanism, auto-encoder can pay more attention to the representation of inlier samples through adversarial training. Secondly, we apply the information entropy into the latent layer to make it sparse and constrain the expression of diversity. Experimental results on three public datasets show that the proposed method achieves comparable performance compared with previous popular approaches.