Fast Particle-based Anomaly Detection Algorithm with Variational Autoencoder
This work addresses anomaly detection for particle physics, specifically in the search for new physics, with incremental improvements in efficiency and latency.
The paper tackles anomaly detection in particle physics by introducing Set-VAE, a particle-based variational autoencoder algorithm, achieving a 2x signal efficiency gain over traditional methods, and proposes CLIP-VAE to reduce inference-time costs with a 2x acceleration in latency.
Model-agnostic anomaly detection is one of the promising approaches in the search for new beyond the standard model physics. In this paper, we present Set-VAE, a particle-based variational autoencoder (VAE) anomaly detection algorithm. We demonstrate a 2x signal efficiency gain compared with traditional subjettiness-based jet selection. Furthermore, with an eye to the future deployment to trigger systems, we propose the CLIP-VAE, which reduces the inference-time cost of anomaly detection by using the KL-divergence loss as the anomaly score, resulting in a 2x acceleration in latency and reducing the caching requirement.