# Challenges for Unsupervised Anomaly Detection in Particle Physics

arXiv: https://arxiv.org/abs/2110.06948 · HTML: https://mlsanity.com/paper/2110.06948/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 94 references that are in the map.
- First posted on arXiv 13 Oct 2021.

## Builds on (most cited first, 30 of 31)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Variational Inference with Normalizing Flows](https://mlsanity.com/paper/1505.05770/) (2015)
- [Variational Inference: A Review for Statisticians](https://mlsanity.com/paper/1601.00670/) (2016)
- [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)](https://mlsanity.com/paper/1511.07289/) (2015)
- [An Introduction to Variational Autoencoders](https://mlsanity.com/paper/1906.02691/) (2019)
- [Improving Variational Inference with Inverse Autoregressive Flow](https://mlsanity.com/paper/1606.04934/) (2016)
- [Sylvester Normalizing Flows for Variational Inference](https://mlsanity.com/paper/1803.05649/) (2018)
- [Anomaly Detection with Density Estimation](https://mlsanity.com/paper/2001.04990/) (2020)
- [Anomaly Detection With Conditional Variational Autoencoders](https://mlsanity.com/paper/2010.05531/) (2020)
- [Variational Autoencoders for New Physics Mining at the Large Hadron Collider](https://mlsanity.com/paper/1811.10276/) (2018)
- [Variational Autoencoders for Anomalous Jet Tagging](https://mlsanity.com/paper/2007.01850/) (2020)
- [Autoencoders for unsupervised anomaly detection in high energy physics](https://mlsanity.com/paper/2104.09051/) (2021)
- [The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider](https://mlsanity.com/paper/2105.14027/) (2021)
- [Simulation Assisted Likelihood-free Anomaly Detection](https://mlsanity.com/paper/2001.05001/) (2020)
- [Weakly Supervised Classification in High Energy Physics](https://mlsanity.com/paper/1702.00414/) (2017)
- [Better Latent Spaces for Better Autoencoders](https://mlsanity.com/paper/2104.08291/) (2021)
- [Novelty Detection Meets Collider Physics](https://mlsanity.com/paper/1807.10261/) (2018)
- [Topological Obstructions to Autoencoding](https://mlsanity.com/paper/2102.08380/) (2021)
- [Bump Hunting in Latent Space](https://mlsanity.com/paper/2103.06595/) (2021)
- [Adversarially Learned Anomaly Detection on CMS Open Data: re-discovering the top quark](https://mlsanity.com/paper/2005.01598/) (2020)
- [Linearized Optimal Transport for Collider Events](https://mlsanity.com/paper/2008.08604/) (2020)
- [Variational Autoencoders for Jet Simulation](https://mlsanity.com/paper/2009.04842/) (2020)
- [Simulation-Assisted Decorrelation for Resonant Anomaly Detection](https://mlsanity.com/paper/2009.02205/) (2020)
- [Comparing Weak- and Unsupervised Methods for Resonant Anomaly Detection](https://mlsanity.com/paper/2104.02092/) (2021)
- [Unsupervised in-distribution anomaly detection of new physics through conditional density estimation](https://mlsanity.com/paper/2012.11638/) (2020)
- [Lund jet images from generative and cycle-consistent adversarial networks](https://mlsanity.com/paper/1909.01359/) (2019)
- [Anomaly Awareness](https://mlsanity.com/paper/2007.14462/) (2020)
- [Learning to Classify from Impure Samples with High-Dimensional Data](https://mlsanity.com/paper/1801.10158/) (2018)
- [An Exploration of Learnt Representations of W Jets](https://mlsanity.com/paper/2109.10919/) (2021)

## Built on by (most cited first, 16 of 16)

- [On Diffusion Modeling for Anomaly Detection](https://mlsanity.com/paper/2305.18593/) (2023)
- [Machine Learning in the Search for New Fundamental Physics](https://mlsanity.com/paper/2112.03769/) (2021)
- [Deep Generative Models for Detector Signature Simulation: A Taxonomic Review](https://mlsanity.com/paper/2312.09597/) (2023)
- [Online-compatible Unsupervised Non-resonant Anomaly Detection](https://mlsanity.com/paper/2111.06417/) (2021)
- [Anomalies, Representations, and Self-Supervision](https://mlsanity.com/paper/2301.04660/) (2023)
- [Snowmass 2021 Computational Frontier CompF03 Topical Group Report: Machine Learning](https://mlsanity.com/paper/2209.07559/) (2022)
- [Improving new physics searches with diffusion models for event observables and jet constituents](https://mlsanity.com/paper/2312.10130/) (2023)
- [Neural Embedding: Learning the Embedding of the Manifold of Physics Data](https://mlsanity.com/paper/2208.05484/) (2022)
- ["Flux+Mutability": A Conditional Generative Approach to One-Class Classification and Anomaly Detection](https://mlsanity.com/paper/2204.08609/) (2022)
- [Machine-Learning Compression for Particle Physics Discoveries](https://mlsanity.com/paper/2210.11489/) (2022)
- [Triggering Dark Showers with Conditional Dual Auto-Encoders](https://mlsanity.com/paper/2306.12955/) (2023)
- [Unsupervised Anomaly Detection via Nonlinear Manifold Learning](https://mlsanity.com/paper/2306.09441/) (2023)
- [AI-based particle track identification in scintillating fibres read out with imaging sensors](https://mlsanity.com/paper/2410.10519/) (2024)
- [TRANSIT your events into a new mass: Fast background interpolation for weakly-supervised anomaly searches](https://mlsanity.com/paper/2503.04342/) (2025)
- [Quantum-Inspired Tensor Network Autoencoders for Anomaly Detection: A MERA-Based Approach](https://mlsanity.com/paper/2604.06541/) (2026)
- [Masked Diffusion Modeling for Anomaly Detection](https://mlsanity.com/paper/2605.30046/) (2026)

## Abstract (arXiv)

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.