Challenges for Unsupervised Anomaly Detection in Particle Physics

Abstract

Anomaly detection relies on designing a score to determine whether a particular event is uncharacteristic of a given background distribution. One way to define a score is to use autoencoders, which rely on the ability to reconstruct certain types of data (background) but not others (signals). In this paper, we study some challenges associated with variational autoencoders, such as the dependence on hyperparameters and the metric used, in the context of anomalous signal (top and $W$) jets in a QCD background. We find that the hyperparameter choices strongly affect the network performance and that the optimal parameters for one signal are non-optimal for another. In exploring the networks, we uncover a connection between the latent space of a variational autoencoder trained using mean-squared-error and the optimal transport distances within the dataset. We then show that optimal transport distances to representative events in the background dataset can be used directly for anomaly detection, with performance comparable to the autoencoders. Whether using autoencoders or optimal transport distances for anomaly detection, we find that the choices that best represent the background are not necessarily best for signal identification. These challenges with unsupervised anomaly detection bolster the case for additional exploration of semi-supervised or alternative approaches.

22 + 2 pages, 8 figures, 2 tables

References in corpus (12)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• An Introduction to PYTHIA 8.2
• Extending the Bump Hunt with Machine Learning
• A generic anti-QCD jet tagger
• Rare and Different: Anomaly Scores from a combination of likelihood and out-of-distribution models to detect new physics at the LHC
• Deep Set Auto Encoders for Anomaly Detection in Particle Physics
• Transferability of Deep Learning Models in Searches for New Physics at Colliders
• Unsupervised in-distribution anomaly detection of new physics through conditional density estimation
• High-dimensional Anomaly Detection with Radiative Return in $e^{+}{e}^{-}$ Collisions
• Uncovering hidden patterns in collider events with Bayesian probabilistic models
• Variational Autoencoders for Jet Simulation
• An Exploration of Learnt Representations of W Jets

Cited by in corpus (20)

• Anomaly detection in high-energy physics using a quantum autoencoder
• Online-compatible Unsupervised Non-resonant Anomaly Detection
• Quantum Anomaly Detection for Collider Physics
• Deep Generative Models for Detector Signature Simulation: A Taxonomic Review
• Variational quantum one-class classifier
• What's Anomalous in LHC Jets?
• Anomaly Detection under Coordinate Transformations
• Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays
• A semi-supervised approach to dark matter searches in direct detection data with machine learning
• Quantum Generative Adversarial Networks For Anomaly Detection In High Energy Physics
• Neural Embedding: Learning the Embedding of the Manifold of Physics Data
• Anomalies, Representations, and Self-Supervision
• Event-based anomaly detection for new physics searches at the LHC using machine learning
• Quantum support vector data description for anomaly detection
• Non-resonant Anomaly Detection with Background Extrapolation
• Unsupervised and lightly supervised learning in particle physics
• Simulation-based Anomaly Detection for Multileptons at the LHC
• Triggering Dark Showers with Conditional Dual Auto-Encoders
• Graph theory inspired anomaly detection at the LHC
• Creating Simple, Interpretable Anomaly Detectors for New Physics in Jet Substructure