Aspen Open Jets: Unlocking LHC Data for Foundation Models in Particle Physics
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914136163614720 |
|---|---|
| author | Amram, Oz Anzalone, Luca Birk, Joschka Faroughy, Darius A. Hallin, Anna Kasieczka, Gregor Krämer, Michael Pang, Ian Reyes-Gonzalez, Humberto Shih, David |
| author_facet | Amram, Oz Anzalone, Luca Birk, Joschka Faroughy, Darius A. Hallin, Anna Kasieczka, Gregor Krämer, Michael Pang, Ian Reyes-Gonzalez, Humberto Shih, David |
| contents | Foundation models are deep learning models pre-trained on large amounts of data which are capable of generalizing to multiple datasets and/or downstream tasks. This work demonstrates how data collected by the CMS experiment at the Large Hadron Collider can be useful in pre-training foundation models for HEP. Specifically, we introduce the AspenOpenJets dataset, consisting of approximately 178M high $p_T$ jets derived from CMS 2016 Open Data. We show how pre-training the OmniJet-$α$ foundation model on AspenOpenJets improves performance on generative tasks with significant domain shift: generating boosted top and QCD jets from the simulated JetClass dataset. In addition to demonstrating the power of pre-training of a jet-based foundation model on actual proton-proton collision data, we provide the ML-ready derived AspenOpenJets dataset for further public use. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_10504 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Aspen Open Jets: Unlocking LHC Data for Foundation Models in Particle Physics Amram, Oz Anzalone, Luca Birk, Joschka Faroughy, Darius A. Hallin, Anna Kasieczka, Gregor Krämer, Michael Pang, Ian Reyes-Gonzalez, Humberto Shih, David High Energy Physics - Phenomenology Machine Learning High Energy Physics - Experiment Foundation models are deep learning models pre-trained on large amounts of data which are capable of generalizing to multiple datasets and/or downstream tasks. This work demonstrates how data collected by the CMS experiment at the Large Hadron Collider can be useful in pre-training foundation models for HEP. Specifically, we introduce the AspenOpenJets dataset, consisting of approximately 178M high $p_T$ jets derived from CMS 2016 Open Data. We show how pre-training the OmniJet-$α$ foundation model on AspenOpenJets improves performance on generative tasks with significant domain shift: generating boosted top and QCD jets from the simulated JetClass dataset. In addition to demonstrating the power of pre-training of a jet-based foundation model on actual proton-proton collision data, we provide the ML-ready derived AspenOpenJets dataset for further public use. |
| title | Aspen Open Jets: Unlocking LHC Data for Foundation Models in Particle Physics |
| topic | High Energy Physics - Phenomenology Machine Learning High Energy Physics - Experiment |
| url | https://arxiv.org/abs/2412.10504 |