Classifying Anomalies THrough Outer Density Estimation (CATHODE)

Abstract

We propose a new model-agnostic search strategy for physics beyond the standard model (BSM) at the LHC, based on a novel application of neural density estimation to anomaly detection. Our approach, which we call Classifying Anomalies THrough Outer Density Estimation (CATHODE), assumes the BSM signal is localized in a signal region (defined e.g. using invariant mass). By training a conditional density estimator on a collection of additional features outside the signal region, interpolating it into the signal region, and sampling from it, we produce a collection of events that follow the background model. We can then train a classifier to distinguish the data from the events sampled from the background model, thereby approaching the optimal anomaly detector. Using the LHC Olympics R&D dataset, we demonstrate that CATHODE nearly saturates the best possible performance, and significantly outperforms other approaches that aim to enhance the bump hunt (CWoLa Hunting and ANODE). Finally, we demonstrate that CATHODE is very robust against correlations between the features and maintains nearly-optimal performance even in this more challenging setting.

arXiv comment17 pages, 12 figures; v2: minor updates; v3 (published version): added study of background sculpting and minor fixes

References in corpus (65)

• Adam: A Method for Stochastic Optimization
• Scikit-learn: Machine Learning in Python
• PYTHIA 6.4 Physics and Manual
• PYTHIA 6.2 Physics and Manual
• The anti-k_t jet clustering algorithm
• A Brief Introduction to PYTHIA 8.1
• FastJet user manual
• DELPHES 3, A modular framework for fast simulation of a generic collider experiment
• Dispelling the N^3 myth for the Kt jet-finder
• Normalizing Flows: An Introduction and Review of Current Methods
• Identifying Boosted Objects with N-subjettiness
• Maximizing Boosted Top Identification by Minimizing N-subjettiness
• Anomaly Detection with Density Estimation
• Classification without labels: Learning from mixed samples in high energy physics
• Searching for New Physics with Deep Autoencoders
• Anomaly Detection for Resonant New Physics with Machine Learning
• Learning New Physics from a Machine
• Extending the Bump Hunt with Machine Learning
• QCD or What?
• A generic anti-QCD jet tagger
• Simulation Assisted Likelihood-free Anomaly Detection
• Novelty Detection Meets Collider Physics
• Adversarially-trained autoencoders for robust unsupervised new physics searches
• Dijet resonance search with weak supervision using $\sqrt{s}=13$ TeV $pp$ collisions in the ATLAS detector
• Guiding New Physics Searches with Unsupervised Learning
• Variational Autoencoders for New Physics Mining at the Large Hadron Collider
• The Dark Machines Anomaly Score Challenge: Benchmark Data and Model Independent Event Classification for the Large Hadron Collider
• Uncovering latent jet substructure
• Tag N' Train: A Technique to Train Improved Classifiers on Unlabeled Data
• Autoencoders for unsupervised anomaly detection in high energy physics
• Anomaly detection with Convolutional Graph Neural Networks
• Global Search for New Physics with 2.0/fb at CDF
• Finding New Physics without learning about it: Anomaly Detection as a tool for Searches at Colliders
• A Living Review of Machine Learning for Particle Physics
• Better Latent Spaces for Better Autoencoders
• Variational Autoencoders for Anomalous Jet Tagging
• Simulation-Assisted Decorrelation for Resonant Anomaly Detection
• GANplifying Event Samples
• A robust anomaly finder based on autoencoders
• Bump Hunting in Latent Space
• Model-Independent and Quasi-Model-Independent Search for New Physics at CDF
• Search for New Physics in e mu X Data at D0 Using Sleuth: A Quasi-Model-Independent Search Strategy for New Physics
• A general search for new phenomena in ep scattering at HERA
• Rare and Different: Anomaly Scores from a combination of likelihood and out-of-distribution models to detect new physics at the LHC
• A Quasi-Model-Independent Search for New Physics at Large Transverse Momentum
• Learning the latent structure of collider events
• Quasi Anomalous Knowledge: Searching for new physics with embedded knowledge
• Unsupervised clustering for collider physics
• A General Search for New Phenomena at HERA
• Comparing Weak- and Unsupervised Methods for Resonant Anomaly Detection
• Mass Unspecific Supervised Tagging (MUST) for boosted jets
• Transferability of Deep Learning Models in Searches for New Physics at Colliders
• Use of a Generalized Energy Mover's Distance in the Search for Rare Phenomena at Colliders
• A Quasi-Model-Independent Search for New High p_T Physics at DZero
• Via Machinae: Searching for Stellar Streams using Unsupervised Machine Learning
• Unsupervised Outlier Detection in Heavy-Ion Collisions
• Anomalous Jet Identification via Sequence Modeling
• Topological Obstructions to Autoencoding
• MUSiC: a model-unspecific search for new physics in proton-proton collisions at $\sqrt{s}=$ 13 TeV
• Machine Learning Templates for QCD Factorization in the Search for Physics Beyond the Standard Model
• Unsupervised in-distribution anomaly detection of new physics through conditional density estimation
• Decoding Dark Matter Substructure without Supervision
• The motivation and status of two-body resonance decays after the LHC Run 2 and beyond
• RanBox: Anomaly Detection in the Copula Space
• Model-Independent Detection of New Physics Signals Using Interpretable Semi-Supervised Classifier Tests

Cited by in corpus (63)

• Machine Learning and LHC Event Generation
• Machine Learning for Anomaly Detection in Particle Physics
• Resonant anomaly detection without background sculpting
• CURTAINs for your Sliding Window: Constructing Unobserved Regions by Transforming Adjacent Intervals
• EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets
• FETA: Flow-Enhanced Transportation for Anomaly Detection
• Improving Variational Autoencoders for New Physics Detection at the LHC with Normalizing Flows
• How to Understand Limitations of Generative Networks
• Learning new physics efficiently with nonparametric methods
• Self-supervised Anomaly Detection for New Physics
• ν-Flows: Conditional Neutrino Regression
• The MadNIS Reloaded
• The Interplay of Machine Learning--based Resonant Anomaly Detection Methods
• Deep Generative Models for Detector Signature Simulation: A Taxonomic Review
• What's Anomalous in LHC Jets?
• Anomaly Detection under Coordinate Transformations
• Model-agnostic search for dijet resonances with anomalous jet substructure in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• High-dimensional and Permutation Invariant Anomaly Detection
• Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays
• Ephemeral Learning -- Augmenting Triggers with Online-Trained Normalizing Flows
• Precision-Machine Learning for the Matrix Element Method
• A Method to Simultaneously Facilitate All Jet Physics Tasks
• Solving Key Challenges in Collider Physics with Foundation Models
• Anomalies, Representations, and Self-Supervision
• Normalizing Flows for High-Dimensional Detector Simulations
• Combining Resonant and Tail-based Anomaly Detection
• Calorimeter shower superresolution
• Event-based anomaly detection for new physics searches at the LHC using machine learning
• Non-Parametric Data-Driven Background Modelling using Conditional Probabilities
• Differentiable MadNIS-Lite
• Convolutional L2LFlows: Generating Accurate Showers in Highly Granular Calorimeters Using Convolutional Normalizing Flows
• Resonant Anomaly Detection with Multiple Reference Datasets
• Enhancing the hunt for new phenomena in dijet final-states using anomaly detection filters at the High-Luminosity Large Hadron Collider
• Flows for Flows: Morphing one Dataset into another with Maximum Likelihood Estimation
• Model selection and signal extraction using Gaussian Process regression
• Anomaly detection with flow-based fast calorimeter simulators
• CaloChallenge 2022: A Community Challenge for Fast Calorimeter Simulation
• Non-resonant Anomaly Detection with Background Extrapolation
• Calibrating Bayesian Generative Machine Learning for Bayesiamplification
• Reweighting simulated events using machine-learning techniques in the CMS experiment
• Unsupervised and lightly supervised learning in particle physics
• Simulation-based Anomaly Detection for Multileptons at the LHC
• Invariant Representation Driven Neural Classifier for Anti-QCD Jet Tagging
• TopicFlow: Disentangling quark and gluon jets with normalizing flows
• Isolating Unisolated Upsilons with Anomaly Detection in CMS Open Data
• Reconstructing axion-like particles from beam dumps with simulation-based inference
• Unifying Simulation and Inference with Normalizing Flows
• Quark-versus-gluon tagging in CMS Open Data with CWoLa and TopicFlow
• Quirk SUEP
• Weakly supervised anomaly detection for resonant new physics in the dijet final state using proton-proton collisions at $\sqrt{s}=13$ TeV with the ATLAS detector
• Multiboson signals in the UN2HDM
• Robust Anomaly Detection for Particle Physics Using Multi-Background Representation Learning
• Finding excesses in model parameter space
• Inferring correlated distributions: boosted top jets
• Data-Driven High-Dimensional Statistical Inference with Generative Models
• Observable Optimization for Precision Theory: Machine Learning Energy Correlators
• OmniLearned: A Foundation Model Framework for All Tasks Involving Jet Physics
• Data driven background estimation in HEP using Generative Adversarial Networks
• Trials Factor for Semi-Supervised NN Classifiers in Searches for Narrow Resonances at the LHC
• Weakly supervised anomaly detection with event-level variables
• Graph theory inspired anomaly detection at the LHC
• Enhancing generalization in high energy physics using white-box adversarial attacks
• Strong CWoLa: Binary Classification Without Background Simulation