Wasserstein normalized autoencoder for anomaly detection

Abstract

A novel anomaly detection algorithm is presented. The Wasserstein normalized autoencoder (WNAE) is a normalized probabilistic model that minimizes the Wasserstein distance between the learned probability distribution--a Boltzmann distribution where the energy is the reconstruction error of the autoencoder--and the distribution of the training data. This algorithm has been developed and applied to the identification of semivisible jets--conical sprays of visible standard model particles and invisible dark matter states--with the CMS experiment at the CERN LHC. Trained on jets of particles from simulated standard model processes, the WNAE is shown to learn the probability distribution of the input data in a fully unsupervised fashion, such that it effectively identifies new physics jets as anomalies. The model exhibits stable, convergent training and recovers strong classification performance for a wide range of signals against the selected background process, for which a standard autoencoder fails because of outlier reconstruction. In addition, the model improves upon standard normalized autoencoders while remaining fully agnostic to the signal. The WNAE directly tackles the problem of outlier reconstruction, a common failure mode of autoencoders in anomaly detection tasks.

Replaced with the published version. Added the journal reference and the DOI. All the figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/MLG-24-002 (CMS Public Pages)

References in corpus (32)

• The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations
• The anti-k_t jet clustering algorithm
• An Introduction to PYTHIA 8.2
• FastJet user manual
• A direct empirical proof of the existence of dark matter
• Parton distributions from high-precision collider data
• Particle-flow reconstruction and global event description with the CMS detector
• Identifying Boosted Objects with N-subjettiness
• Echoes of a Hidden Valley at Hadron Colliders
• The CMS trigger system
• Soft Drop
• Description and performance of track and primary-vertex reconstruction with the CMS tracker
• Jet energy scale and resolution in the CMS experiment in pp collisions at 8 TeV
• Performance of the CMS muon detector and muon reconstruction with proton-proton collisions at $\sqrt{s}=$ 13 TeV
• Energy Correlation Functions for Jet Substructure
• Performance of the CMS Level-1 trigger in proton-proton collisions at $\sqrt{s}=$ 13 TeV
• Electron and photon reconstruction and identification with the CMS experiment at the CERN LHC
• Searching for New Physics with Deep Autoencoders
• Performance of missing transverse momentum reconstruction in proton-proton collisions at $\sqrt{s}=$ 13 TeV using the CMS detector
• Semi-visible Jets: Dark Matter Undercover at the LHC
• QCD or What?
• Energy flow polynomials: A complete linear basis for jet substructure
• LHC Searches for Dark Sector Showers
• Autoencoders for unsupervised anomaly detection in high energy physics
• Dark Energy Survey Year 1 Results: Curved-Sky Weak Lensing Mass Map
• Strongly interacting dark sectors in the early Universe and at the LHC through a simplified portal
• Autoencoders for Semivisible Jet Detection
• Performance of the CMS high-level trigger during LHC Run 2
• Search for resonant production of strongly coupled dark matter in proton-proton collisions at 13 TeV
• Search for new phenomena in two-body invariant mass distributions using unsupervised machine learning for anomaly detection at $\sqrt{s}=13$ TeV with the ATLAS detector
• Model-agnostic search for dijet resonances with anomalous jet substructure in proton-proton collisions at $\sqrt{s}$ = 13 TeV
• Search for new physics in final states with semi-visible jets or anomalous signatures using the ATLAS detector