Autoencoders for unsupervised anomaly detection in high energy physics

Abstract

Autoencoders are widely used in machine learning applications, in particular for anomaly detection. Hence, they have been introduced in high energy physics as a promising tool for model-independent new physics searches. We scrutinize the usage of autoencoders for unsupervised anomaly detection based on reconstruction loss to show their capabilities, but also their limitations. As a particle physics benchmark scenario, we study the tagging of top jet images in a background of QCD jet images. Although we reproduce the positive results from the literature, we show that the standard autoencoder setup cannot be considered as a model-independent anomaly tagger by inverting the task: due to the sparsity and the specific structure of the jet images, the autoencoder fails to tag QCD jets if it is trained on top jets even in a semi-supervised setup. Since the same autoencoder architecture can be a good tagger for a specific example of an anomaly and a bad tagger for a different example, we suggest improved performance measures for the task of model-independent anomaly detection. We also improve the capability of the autoencoder to learn non-trivial features of the jet images, such that it is able to achieve both top jet tagging and the inverse task of QCD jet tagging with the same setup. However, we want to stress that a truly model-independent and powerful autoencoder-based unsupervised jet tagger still needs to be developed.

arXiv comment32 pages, 20 figures

References in corpus (30)

• The anti-k_t jet clustering algorithm
• An Introduction to PYTHIA 8.2
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Deep Learning for Anomaly Detection: A Review
• A Unifying Review of Deep and Shallow Anomaly Detection
• Deep Learning and its Application to LHC Physics
• Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
• Classification without labels: Learning from mixed samples in high energy physics
• Deep-learning Top Taggers or The End of QCD?
• Searching for New Physics with Deep Autoencoders
• Deep-learned Top Tagging with a Lorentz Layer
• How to GAN LHC Events
• QCD or What?
• Novelty Detection Meets Collider Physics
• Adversarially-trained autoencoders for robust unsupervised new physics searches
• Machine and Deep Learning Applications in Particle Physics
• Pulling Out All the Tops with Computer Vision and Deep Learning
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• Learning to Classify from Impure Samples with High-Dimensional Data
• Tag N' Train: A Technique to Train Improved Classifiers on Unlabeled Data
• Weakly Supervised Classification in High Energy Physics
• Finding New Physics without learning about it: Anomaly Detection as a tool for Searches at Colliders
• (Machine) Learning to Do More with Less
• Casting a graph net to catch dark showers
• Learning the latent structure of collider events
• Topological Obstructions to Autoencoding
• Combine and Conquer: Event Reconstruction with Bayesian Ensemble Neural Networks
• Quark Gluon Jet Discrimination with Weakly Supervised Learning
• Machine Learning on data with sPlot background subtraction

Cited by in corpus (37)

• Anomaly detection in high-energy physics using a quantum autoencoder
• Classifying Anomalies THrough Outer Density Estimation (CATHODE)
• Anomaly detection with Convolutional Graph Neural Networks
• Bump Hunting in Latent Space
• Machine Learning for Anomaly Detection in Particle Physics
• Challenges for Unsupervised Anomaly Detection in Particle Physics
• Autoencoders for Semivisible Jet Detection
• Online-compatible Unsupervised Non-resonant Anomaly Detection
• Deep Set Auto Encoders for Anomaly Detection in Particle Physics
• Improving Variational Autoencoders for New Physics Detection at the LHC with Normalizing Flows
• Unsupervised Hadronic SUEP at the LHC
• Learning new physics efficiently with nonparametric methods
• Deep Generative Models for Detector Signature Simulation: A Taxonomic Review
• What's Anomalous in LHC Jets?
• Anomaly Detection under Coordinate Transformations
• Anomaly Detection Based on Isolation Mechanisms: A Survey
• Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays
• Neural Embedding: Learning the Embedding of the Manifold of Physics Data
• Knowledge-integrated AutoEncoder Model
• GAN-AE : An anomaly detection algorithm for New Physics search in LHC data
• Searching for dark jets with displaced vertices using weakly supervised machine learning
• Deep learning techniques for Imaging Air Cherenkov Telescopes
• Semi-visible jets, energy-based models, and self-supervision
• Radio-astronomical Image Reconstruction with Conditional Denoising Diffusion Model
• Non-resonant Anomaly Detection with Background Extrapolation
• High-dimensional Anomaly Detection with Radiative Return in $e^{+}{e}^{-}$ Collisions
• Jet substructure observables for jet quenching in Quark Gluon Plasma: a Machine Learning driven analysis
• Invariant Representation Driven Neural Classifier for Anti-QCD Jet Tagging
• Simulation-based Anomaly Detection for Multileptons at the LHC
• Unsupervised and lightly supervised learning in particle physics
• Interplay of Traditional Methods and Machine Learning Algorithms for Tagging Boosted Objects
• Generative Invertible Quantum Neural Networks
• Triggering Dark Showers with Conditional Dual Auto-Encoders
• Detecting New Physics as Novelty -- Complementarity Matters
• Wasserstein normalized autoencoder for anomaly detection
• Graph theory inspired anomaly detection at the LHC
• Creating Simple, Interpretable Anomaly Detectors for New Physics in Jet Substructure