LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows
In high energy physics, one of the most important processes for collider data analysis is the comparison of collected and simulated data. Nowadays the state-of-the-art for data generation is in the form of Monte Carlo (MC) generators. However, because of the upcoming high-luminosity upgrade of the L...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | Machine Learning: Science and Technology |
| Subjects: | |
| Online Access: | https://doi.org/10.1088/2632-2153/ad04ea |
| _version_ | 1797644345810092032 |
|---|---|
| author | Breno Orzari Nadezda Chernyavskaya Raphael Cobe Javier Duarte Jefferson Fialho Dimitrios Gunopulos Raghav Kansal Maurizio Pierini Thiago Tomei Mary Touranakou |
| author_facet | Breno Orzari Nadezda Chernyavskaya Raphael Cobe Javier Duarte Jefferson Fialho Dimitrios Gunopulos Raghav Kansal Maurizio Pierini Thiago Tomei Mary Touranakou |
| author_sort | Breno Orzari |
| collection | DOAJ |
| description | In high energy physics, one of the most important processes for collider data analysis is the comparison of collected and simulated data. Nowadays the state-of-the-art for data generation is in the form of Monte Carlo (MC) generators. However, because of the upcoming high-luminosity upgrade of the Large Hadron Collider (LHC), there will not be enough computational power or time to match the amount of needed simulated data using MC methods. An alternative approach under study is the usage of machine learning generative methods to fulfill that task. Since the most common final-state objects of high-energy proton collisions are hadronic jets, which are collections of particles collimated in a given region of space, this work aims to develop a convolutional variational autoencoder (ConVAE) for the generation of particle-based LHC hadronic jets. Given the ConVAE’s limitations, a normalizing flow (NF) network is coupled to it in a two-step training process, which shows improvements on the results for the generated jets. The ConVAE+NF network is capable of generating a jet in $18.30 \pm 0.04\,\,{\mu\text{s}}$ , making it one of the fastest methods for this task up to now. |
| first_indexed | 2024-03-11T14:30:14Z |
| format | Article |
| id | doaj.art-75b47c62b99d4ebaa1d38fe08ae5ec2f |
| institution | Directory Open Access Journal |
| issn | 2632-2153 |
| language | English |
| last_indexed | 2024-03-11T14:30:14Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | Machine Learning: Science and Technology |
| spelling | doaj.art-75b47c62b99d4ebaa1d38fe08ae5ec2f2023-10-31T10:17:39ZengIOP PublishingMachine Learning: Science and Technology2632-21532023-01-014404502310.1088/2632-2153/ad04eaLHC hadronic jet generation using convolutional variational autoencoders with normalizing flowsBreno Orzari0https://orcid.org/0000-0003-4232-4743Nadezda Chernyavskaya1https://orcid.org/0000-0002-2264-2229Raphael Cobe2https://orcid.org/0000-0002-0852-2183Javier Duarte3https://orcid.org/0000-0002-5076-7096Jefferson Fialho4https://orcid.org/0000-0002-5421-0789Dimitrios Gunopulos5https://orcid.org/0000-0001-6339-1879Raghav Kansal6https://orcid.org/0000-0003-2445-1060Maurizio Pierini7https://orcid.org/0000-0003-1939-4268Thiago Tomei8https://orcid.org/0000-0002-1809-5226Mary Touranakou9https://orcid.org/0000-0002-3682-3258Universidade Estadual Paulista , São Paulo/SP-CEP 01049-010, BrazilEuropean Organization for Nuclear Research (CERN) , CH-1211 Geneva 23, SwitzerlandUniversidade Estadual Paulista , São Paulo/SP-CEP 01049-010, BrazilUniversity of California San Diego , La Jolla, CA 92093, United States of AmericaUniversidade Estadual Paulista , São Paulo/SP-CEP 01049-010, BrazilDepartment of Informatics and Telecommunications, National and Kapodistrian University of Athens , Athens 157 72, GreeceUniversity of California San Diego , La Jolla, CA 92093, United States of AmericaEuropean Organization for Nuclear Research (CERN) , CH-1211 Geneva 23, SwitzerlandUniversidade Estadual Paulista , São Paulo/SP-CEP 01049-010, BrazilEuropean Organization for Nuclear Research (CERN) , CH-1211 Geneva 23, Switzerland; Department of Informatics and Telecommunications, National and Kapodistrian University of Athens , Athens 157 72, GreeceIn high energy physics, one of the most important processes for collider data analysis is the comparison of collected and simulated data. Nowadays the state-of-the-art for data generation is in the form of Monte Carlo (MC) generators. However, because of the upcoming high-luminosity upgrade of the Large Hadron Collider (LHC), there will not be enough computational power or time to match the amount of needed simulated data using MC methods. An alternative approach under study is the usage of machine learning generative methods to fulfill that task. Since the most common final-state objects of high-energy proton collisions are hadronic jets, which are collections of particles collimated in a given region of space, this work aims to develop a convolutional variational autoencoder (ConVAE) for the generation of particle-based LHC hadronic jets. Given the ConVAE’s limitations, a normalizing flow (NF) network is coupled to it in a two-step training process, which shows improvements on the results for the generated jets. The ConVAE+NF network is capable of generating a jet in $18.30 \pm 0.04\,\,{\mu\text{s}}$ , making it one of the fastest methods for this task up to now.https://doi.org/10.1088/2632-2153/ad04eagenerative modelsparticle physicshigh energy physicshyperparameter tuning |
| spellingShingle | Breno Orzari Nadezda Chernyavskaya Raphael Cobe Javier Duarte Jefferson Fialho Dimitrios Gunopulos Raghav Kansal Maurizio Pierini Thiago Tomei Mary Touranakou LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows Machine Learning: Science and Technology generative models particle physics high energy physics hyperparameter tuning |
| title | LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| title_full | LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| title_fullStr | LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| title_full_unstemmed | LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| title_short | LHC hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| title_sort | lhc hadronic jet generation using convolutional variational autoencoders with normalizing flows |
| topic | generative models particle physics high energy physics hyperparameter tuning |
| url | https://doi.org/10.1088/2632-2153/ad04ea |
| work_keys_str_mv | AT brenoorzari lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT nadezdachernyavskaya lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT raphaelcobe lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT javierduarte lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT jeffersonfialho lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT dimitriosgunopulos lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT raghavkansal lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT mauriziopierini lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT thiagotomei lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows AT marytouranakou lhchadronicjetgenerationusingconvolutionalvariationalautoencoderswithnormalizingflows |