Self-supervised anomaly detection for new physics
We investigate a method of model-agnostic anomaly detection through studying jets, collimated sprays of particles produced in high-energy collisions. We train a transformer neural network to encode simulated QCD “event space” dijets into a low-dimensional “latent space” representation. We optimize t...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
8 September 2022
|
| In: |
Physical review
Year: 2022, Jahrgang: 106, Heft: 5, Pages: 1-12 |
| ISSN: | 2470-0029 |
| DOI: | 10.1103/PhysRevD.106.056005 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1103/PhysRevD.106.056005 Verlag, kostenfrei, Volltext: https://link.aps.org/doi/10.1103/PhysRevD.106.056005 
|
| Verfasserangaben: | Barry M. Dillon, Radha Mastandrea, and Benjamin Nachman |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 183370584X | ||
| 003 | DE-627 | ||
| 005 | 20230706234011.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 230208s2022 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1103/PhysRevD.106.056005 |2 doi | |
| 035 | |a (DE-627)183370584X | ||
| 035 | |a (DE-599)KXP183370584X | ||
| 035 | |a (OCoLC)1389535803 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 29 |2 sdnb | ||
| 100 | 1 | |a Dillon, Barry M. |e VerfasserIn |0 (DE-588)1243203412 |0 (DE-627)1773856952 |4 aut | |
| 245 | 1 | 0 | |a Self-supervised anomaly detection for new physics |c Barry M. Dillon, Radha Mastandrea, and Benjamin Nachman |
| 264 | 1 | |c 8 September 2022 | |
| 300 | |a 12 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 08.02.2023 | ||
| 520 | |a We investigate a method of model-agnostic anomaly detection through studying jets, collimated sprays of particles produced in high-energy collisions. We train a transformer neural network to encode simulated QCD “event space” dijets into a low-dimensional “latent space” representation. We optimize the network using the self-supervised contrastive loss, which encourages the preservation of known physical symmetries of the dijets. We then train a binary classifier to discriminate a beyond the standard model resonant dijet signal from a QCD dijet background both in the event space and the latent space representations. We find the classifier performances on the event and latent spaces to be comparable. We finally perform an anomaly detection search using a weakly supervised bump hunt on the latent space dijets, finding again a comparable performance to a search run on the physical space dijets. This opens the door to using low-dimensional latent representations as a computationally efficient space for resonant anomaly detection in generic particle collision events. | ||
| 700 | 1 | |a Mastandrea, Radha |e VerfasserIn |0 (DE-588)1280526939 |0 (DE-627)1833782798 |4 aut | |
| 700 | 1 | |a Nachman, Benjamin |e VerfasserIn |0 (DE-588)1280527072 |0 (DE-627)1833783190 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Physical review |d Ridge, NY : American Physical Society, 2016 |g 106(2022), 5, Artikel-ID 056005, Seite 1-12 |h Online-Ressource |w (DE-627)846313510 |w (DE-600)2844732-3 |w (DE-576)454495811 |x 2470-0029 |7 nnas |a Self-supervised anomaly detection for new physics |
| 773 | 1 | 8 | |g volume:106 |g year:2022 |g number:5 |g elocationid:056005 |g pages:1-12 |g extent:12 |a Self-supervised anomaly detection for new physics |
| 856 | 4 | 0 | |u https://doi.org/10.1103/PhysRevD.106.056005 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u https://link.aps.org/doi/10.1103/PhysRevD.106.056005 |x Verlag |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20230208 | ||
| 993 | |a Article | ||
| 994 | |a 2022 | ||
| 998 | |g 1243203412 |a Dillon, Barry M. |m 1243203412:Dillon, Barry M. |d 130000 |d 130300 |e 130000PD1243203412 |e 130300PD1243203412 |k 0/130000/ |k 1/130000/130300/ |p 1 |x j | ||
| 999 | |a KXP-PPN183370584X |e 4268591435 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"title":[{"title_sort":"Self-supervised anomaly detection for new physics","title":"Self-supervised anomaly detection for new physics"}],"person":[{"given":"Barry M.","family":"Dillon","role":"aut","display":"Dillon, Barry M.","roleDisplay":"VerfasserIn"},{"family":"Mastandrea","given":"Radha","roleDisplay":"VerfasserIn","display":"Mastandrea, Radha","role":"aut"},{"role":"aut","roleDisplay":"VerfasserIn","display":"Nachman, Benjamin","given":"Benjamin","family":"Nachman"}],"note":["Gesehen am 08.02.2023"],"type":{"media":"Online-Ressource","bibl":"article-journal"},"language":["eng"],"recId":"183370584X","origin":[{"dateIssuedKey":"2022","dateIssuedDisp":"8 September 2022"}],"id":{"eki":["183370584X"],"doi":["10.1103/PhysRevD.106.056005"]},"name":{"displayForm":["Barry M. Dillon, Radha Mastandrea, and Benjamin Nachman"]},"physDesc":[{"extent":"12 S."}],"relHost":[{"title":[{"title_sort":"Physical review","title":"Physical review"}],"pubHistory":["3rd series, volume 93, number 1 (January 2016)-"],"titleAlt":[{"title":"Particles, fields, gravitation, and cosmology"}],"part":{"volume":"106","text":"106(2022), 5, Artikel-ID 056005, Seite 1-12","extent":"12","year":"2022","issue":"5","pages":"1-12"},"disp":"Self-supervised anomaly detection for new physicsPhysical review","note":["Gesehen am 14.03.2023"],"type":{"media":"Online-Ressource","bibl":"periodical"},"language":["eng"],"corporate":[{"roleDisplay":"Herausgebendes Organ","display":"American Physical Society","role":"isb"}],"recId":"846313510","origin":[{"publisherPlace":"Ridge, NY","dateIssuedDisp":"2016-","dateIssuedKey":"2016","publisher":"American Physical Society"}],"id":{"issn":["2470-0029"],"eki":["846313510"],"zdb":["2844732-3"]},"name":{"displayForm":["published by American Physical Society"]},"physDesc":[{"extent":"Online-Ressource"}]}]} | ||
| SRT | |a DILLONBARRSELFSUPERV8202 | ||