A Caltech Library Service

Improving Variational Autoencoders for New Physics Detection at the LHC With Normalizing Flows

Jawahar, Pratik and Aarrestad, Thea and Chernyavskaya, Nadezda and Pierini, Maurizio and Wozniak, Kinga A. and Ngadiuba, Jennifer and Duarte, Javier and Tsan, Steven (2022) Improving Variational Autoencoders for New Physics Detection at the LHC With Normalizing Flows. Frontiers in Big Data, 5 . Art. No. 803685. ISSN 2624-909X. PMCID PMC8919050. doi:10.3389/fdata.2022.803685.

[img] PDF - Published Version
Creative Commons Attribution.

[img] PDF - Supplemental Material
Creative Commons Attribution.


Use this Persistent URL to link to this item:


We investigate how to improve new physics detection strategies exploiting variational autoencoders and normalizing flows for anomaly detection at the Large Hadron Collider. As a working example, we consider the DarkMachines challenge dataset. We show how different design choices (e.g., event representations, anomaly score definitions, network architectures) affect the result on specific benchmark new physics models. Once a baseline is established, we discuss how to improve the anomaly detection accuracy by exploiting normalizing flow layers in the latent space of the variational autoencoder.

Item Type:Article
Related URLs:
URLURL TypeDescription CentralArticle
Aarrestad, Thea0000-0002-7671-243X
Pierini, Maurizio0000-0003-1939-4268
Ngadiuba, Jennifer0000-0002-0055-2935
Duarte, Javier0000-0002-5076-7096
Additional Information:© 2022 Jawahar, Aarrestad, Chernyavskaya, Pierini, Wozniak, Ngadiuba, Duarte and Tsan. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Received: 28 October 2021; Accepted: 17 January 2022; Published: 28 February 2022. PJ, TA, MP, and KW were supported by the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (Grant Agreement No. 772369). JD was supported by the U.S. Department of Energy (DOE), Office of Science, Office of High Energy Physics Early Career Research program under Award No. DE-SC0021187. ST was supported by the University of California San Diego Triton Research and Experiential Learning Scholars (TRELS) program. JN was supported by Fermi Research Alliance, LLC under Contract No. DE-AC02-07CH11359 with the U.S. Department of Energy, Office of Science, Office of High Energy Physics. Author Contributions. All authors in equal share developed the baseline VAE model. All authors in equal share took part in writing and editing the manuscript. PJ and MP developed the VAE + normalizing flow models. All authors contributed to the article and approved the submitted version. Data Availability Statement. The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. The Supplementary Material for this article can be found online at:
Funding AgencyGrant Number
European Research Council (ERC)772369
Department of Energy (DOE)DE-SC0021187
University of California, San DiegoUNSPECIFIED
Department of Energy (DOE)DE-AC02-07CH11359
Subject Keywords:anomaly detection (AD), variational auto encoder (VAE), normalizing flow (NF), Large Hadron Collider (LHC), new physics beyond standard model, graph convolutional network (GCN), convolutional neural net
PubMed Central ID:PMC8919050
Record Number:CaltechAUTHORS:20220322-742433000
Persistent URL:
Official Citation:Jawahar P, Aarrestad T, Chernyavskaya N, Pierini M, Wozniak KA, Ngadiuba J, Duarte J and Tsan S (2022) Improving Variational Autoencoders for New Physics Detection at the LHC With Normalizing Flows. Front. Big Data 5:803685. doi: 10.3389/fdata.2022.803685
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:114008
Deposited By: George Porter
Deposited On:23 Mar 2022 14:56
Last Modified:23 Mar 2022 14:56

Repository Staff Only: item control page