On Diffusion Modeling for Anomaly Detection
This work addresses anomaly detection for applications requiring scalable and efficient methods, offering a competitive alternative to existing techniques, though it is incremental in adapting diffusion models to this domain.
The paper tackled the problem of anomaly detection using diffusion models, finding that Denoising Diffusion Probability Models (DDPM) perform well but are computationally expensive, and proposed Diffusion Time Estimation (DTE) as a faster alternative that outperforms DDPM on the ADBench benchmark with orders of magnitude faster inference time.
Known for their impressive performance in generative modeling, diffusion models are attractive candidates for density-based anomaly detection. This paper investigates different variations of diffusion modeling for unsupervised and semi-supervised anomaly detection. In particular, we find that Denoising Diffusion Probability Models (DDPM) are performant on anomaly detection benchmarks yet computationally expensive. By simplifying DDPM in application to anomaly detection, we are naturally led to an alternative approach called Diffusion Time Estimation (DTE). DTE estimates the distribution over diffusion time for a given input and uses the mode or mean of this distribution as the anomaly score. We derive an analytical form for this density and leverage a deep neural network to improve inference efficiency. Through empirical evaluations on the ADBench benchmark, we demonstrate that all diffusion-based anomaly detection methods perform competitively for both semi-supervised and unsupervised settings. Notably, DTE achieves orders of magnitude faster inference time than DDPM, while outperforming it on this benchmark. These results establish diffusion-based anomaly detection as a scalable alternative to traditional methods and recent deep-learning techniques for standard unsupervised and semi-supervised anomaly detection settings.