mlsanityData to 7 Oct 2026

Autoencoders for unsupervised anomaly detection in high energy physics

arXiv 2104.09051 ↗First posted 19 Apr 2021hep-ph cs.LG physics.data-ancited by 32

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). Builds on 25 of its 68 references that are in the map. First posted on arXiv 19 Apr 2021.

Builds on

25 of 68 references.

References · 25

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2019Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detectioncited by 560
  3. 2019Likelihood Ratios for Out-of-Distribution Detectioncited by 451
  4. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  5. 2020A Unifying Review of Deep and Shallow Anomaly Detectioncited by 423
  6. 2019Deep Learning for Anomaly Detection: A Surveycited by 357
  7. 2020Why Normalizing Flows Fail to Detect Out-of-Distribution Datacited by 220
  8. 2019Input complexity and out-of-distribution detection with likelihood-based generative modelscited by 186
  9. 2020Deep Learning for Anomaly Detection: A Reviewcited by 159
  10. 2021A Living Review of Machine Learning for Particle Physicscited by 77
  11. 2020Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Featurescited by 64
  12. 2018Machine Learning in High Energy Physics Community White Papercited by 48
13 more
  1. 2017Classification without labels: Learning from mixed samples in high energy physicscited by 42
  2. 2018Variational Autoencoders for New Physics Mining at the Large Hadron Collidercited by 40
  3. 2020Variational Autoencoders for Anomalous Jet Taggingcited by 33
  4. 2017Weakly Supervised Classification in High Energy Physicscited by 21
  5. 2018Novelty Detection Meets Collider Physicscited by 20
  6. 2021Topological Obstructions to Autoencodingcited by 20
  7. 2021Better Latent Spaces for Better Autoencoderscited by 20
  8. 2021Bump Hunting in Latent Spacecited by 19
  9. 2017Jet Constituents for Deep Neural Network Based Top Quark Taggingcited by 15
  10. 2021Comparing Weak- and Unsupervised Methods for Resonant Anomaly Detectioncited by 13
  11. 2018Learning to Classify from Impure Samples with High-Dimensional Datacited by 8
  12. 2017(Machine) Learning to Do More with Lesscited by 7
  13. 2019Machine Learning on data with sPlot background subtractioncited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 7; 2023: 10; 2024: 6; 2025: 4; 2026: 1.

32 papers.

Sort:

By area:

  1. 2021Machine Learning in the Search for New Fundamental Physicscited by 50
  2. 2023Machine Learning for Anomaly Detection in Particle Physicscited by 29
  3. 2022Learning new physics efficiently with nonparametric methodscited by 20
  4. 2023Deep Generative Models for Detector Signature Simulation: A Taxonomic Reviewcited by 18
  5. 2021Challenges for Unsupervised Anomaly Detection in Particle Physicscited by 16
  6. 2021Online-compatible Unsupervised Non-resonant Anomaly Detectioncited by 15
  7. 2022Lorentz group equivariant autoencoderscited by 15
  8. 2022Snowmass 2021 Computational Frontier CompF03 Topical Group Report: Machine Learningcited by 10
  9. 2023Improving new physics searches with diffusion models for event observables and jet constituentscited by 10
  10. 2024Anomaly Detection Based on Isolation Mechanisms: A Surveycited by 10
  11. 2021Autoencoders for Semivisible Jet Detectioncited by 9
  12. 2022IRC-safe Graph Autoencoder for unsupervised anomaly detectioncited by 8
20 more
  1. 2022Neural Embedding: Learning the Embedding of the Manifold of Physics Datacited by 7
  2. 2023Fast Particle-based Anomaly Detection Algorithm with Variational Autoencodercited by 6
  3. 2024Generating Physical Dynamics under Priorscited by 6
  4. 2022Invariant Representation Driven Neural Classifier for Anti-QCD Jet Taggingcited by 4
  5. 2023Comparative Evaluation of Anomaly Detection Methods for Fraud Detection in Online Credit Card Paymentscited by 4
  6. 2024Radio-astronomical Image Reconstruction with Conditional Denoising Diffusion Modelcited by 4
  7. 2023Generative Invertible Quantum Neural Networkscited by 3
  8. 2023AD-NEV: A Scalable Multi-level Neuroevolution Framework for Multivariate Anomaly Detectioncited by 3
  9. 2023Versatile Energy-Based Probabilistic Models for High Energy Physicscited by 2
  10. 2025AutoSciDACT: Automated Scientific Discovery through Contrastive Embedding and Hypothesis Testingcited by 2
  11. 2023Targeted collapse regularized autoencoder for anomaly detection: black hole at the centercited by 1
  12. 2023Triggering Dark Showers with Conditional Dual Auto-Encoderscited by 1
  13. 2024A comprehensive GeoAI review: Progress, Challenges and Outlookscited by 1
  14. 2025Fast reconstruction-based ROI triggering via anomaly detection in the CYGNO optical TPCcited by 1
  15. 2022Radial Autoencoders for Enhanced Anomaly Detectioncited by 0
  16. 2024AD-NEv++ : The multi-architecture neuroevolution-based multivariate anomaly detection frameworkcited by 0
  17. 2024A Data Mining-Based Dynamical Anomaly Detection Method for Integrating with an Advance Metering Systemcited by 0
  18. 2025TRANSIT your events into a new mass: Fast background interpolation for weakly-supervised anomaly searchescited by 0
  19. 2025A Generative Model for Disentangling Galaxy Photometric Parameterscited by 0
  20. 2026Quantum-Inspired Tensor Network Autoencoders for Anomaly Detection: A MERA-Based Approachcited by 0