Normalizing flows for high-dimensional detector simulations

SciPost Journals Publication Detail SciPost Phys. 18, 081 (2025) Normalizing flows for high-dimensional detector simulations

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ernst, Florian (VerfasserIn) , Favaro, Luigi (VerfasserIn) , Krause, Claudius (VerfasserIn) , Plehn, Tilman (VerfasserIn) , Shih, David (VerfasserIn)
Dokumenttyp: Article (Journal)
Sprache:Englisch
Veröffentlicht: 5 March 2025
In: SciPost physics
Year: 2025, Jahrgang: 18, Heft: 3, Pages: 1-33
ISSN:2542-4653
DOI:10.21468/SciPostPhys.18.3.081
Online-Zugang:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.21468/SciPostPhys.18.3.081
Verlag, lizenzpflichtig, Volltext: https://scipost.org/10.21468/SciPostPhys.18.3.081
Volltext
Verfasserangaben:Florian Ernst, Luigi Favaro, Claudius Krause, Tilman Plehn and David Shih

MARC

LEADER 00000naa a2200000 c 4500
001 1935229141
003 DE-627
005 20250903105340.0
007 cr uuu---uuuuu
008 250903s2025 xx |||||o 00| ||eng c
024 7 |a 10.21468/SciPostPhys.18.3.081  |2 doi 
035 |a (DE-627)1935229141 
035 |a (DE-599)KXP1935229141 
040 |a DE-627  |b ger  |c DE-627  |e rda 
041 |a eng 
084 |a 29  |2 sdnb 
100 1 |a Ernst, Florian  |e VerfasserIn  |0 (DE-588)1375392336  |0 (DE-627)1935230069  |4 aut 
245 1 0 |a Normalizing flows for high-dimensional detector simulations  |c Florian Ernst, Luigi Favaro, Claudius Krause, Tilman Plehn and David Shih 
264 1 |c 5 March 2025 
300 |b Illustrationen 
300 |a 33 
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 03.09.2025 
520 |a SciPost Journals Publication Detail SciPost Phys. 18, 081 (2025) Normalizing flows for high-dimensional detector simulations 
520 |a Whenever invertible generative networks are needed for LHC physics, normalizing flows show excellent performance. In this work, we investigate their performance for fast calorimeter shower simulations with increasing phase space dimension. We use fast and expressive coupling spline transformations applied to the CaloChallenge datasets. In addition to the base flow architecture we also employ a VAE to compress the dimensionality and train a generative network in the latent space. We evaluate our networks on several metrics, including high-level features, classifiers, and generation timing. Our findings demonstrate that invertible neural networks have competitive performance when compared to autoregressive flows, while being substantially faster during generation. 
700 1 |a Favaro, Luigi  |d 1996-  |e VerfasserIn  |0 (DE-588)1331722098  |0 (DE-627)1890706329  |4 aut 
700 1 |a Krause, Claudius  |e VerfasserIn  |0 (DE-588)1114848948  |0 (DE-627)869248847  |0 (DE-576)477537456  |4 aut 
700 1 |a Plehn, Tilman  |d 1969-  |e VerfasserIn  |0 (DE-588)1021935573  |0 (DE-627)715839535  |0 (DE-576)363449809  |4 aut 
700 1 |a Shih, David  |e VerfasserIn  |0 (DE-588)134814940X  |0 (DE-627)1908390026  |4 aut 
773 0 8 |i Enthalten in  |t SciPost physics  |d Amsterdam : SciPost Foundation, 2016  |g 18(2025), 3, Artikel-ID 081, Seite 1-33  |h Online-Ressource  |w (DE-627)881391751  |w (DE-600)2886659-9  |w (DE-576)484813447  |x 2542-4653  |7 nnas  |a Normalizing flows for high-dimensional detector simulations 
773 1 8 |g volume:18  |g year:2025  |g number:3  |g elocationid:081  |g pages:1-33  |g extent:33  |a Normalizing flows for high-dimensional detector simulations 
856 4 0 |u https://doi.org/10.21468/SciPostPhys.18.3.081  |x Verlag  |x Resolving-System  |z lizenzpflichtig  |3 Volltext 
856 4 0 |u https://scipost.org/10.21468/SciPostPhys.18.3.081  |x Verlag  |z lizenzpflichtig  |3 Volltext 
951 |a AR 
992 |a 20250903 
993 |a Article 
994 |a 2025 
998 |g 1021935573  |a Plehn, Tilman  |m 1021935573:Plehn, Tilman  |d 130000  |d 130300  |e 130000PP1021935573  |e 130300PP1021935573  |k 0/130000/  |k 1/130000/130300/  |p 4 
998 |g 1331722098  |a Favaro, Luigi  |m 1331722098:Favaro, Luigi  |d 130000  |d 130300  |e 130000PF1331722098  |e 130300PF1331722098  |k 0/130000/  |k 1/130000/130300/  |p 2 
998 |g 1375392336  |a Ernst, Florian  |m 1375392336:Ernst, Florian  |d 130000  |e 130000PE1375392336  |k 0/130000/  |p 1  |x j 
999 |a KXP-PPN1935229141  |e 4767208378 
BIB |a Y 
SER |a journal 
JSO |a {"id":{"eki":["1935229141"],"doi":["10.21468/SciPostPhys.18.3.081"]},"origin":[{"dateIssuedKey":"2025","dateIssuedDisp":"5 March 2025"}],"name":{"displayForm":["Florian Ernst, Luigi Favaro, Claudius Krause, Tilman Plehn and David Shih"]},"relHost":[{"physDesc":[{"extent":"Online-Ressource"}],"id":{"zdb":["2886659-9"],"eki":["881391751"],"issn":["2542-4653"]},"origin":[{"publisher":"SciPost Foundation","dateIssuedDisp":"[2016]-","publisherPlace":"Amsterdam"}],"language":["eng"],"recId":"881391751","type":{"bibl":"periodical","media":"Online-Ressource"},"note":["Gesehen am 06.03.17"],"disp":"Normalizing flows for high-dimensional detector simulationsSciPost physics","part":{"year":"2025","pages":"1-33","issue":"3","volume":"18","text":"18(2025), 3, Artikel-ID 081, Seite 1-33","extent":"33"},"pubHistory":["Vol. 1, issue 1 (September/October 2016)-"],"title":[{"title_sort":"SciPost physics","title":"SciPost physics"}]}],"physDesc":[{"noteIll":"Illustrationen","extent":"33 S."}],"title":[{"title":"Normalizing flows for high-dimensional detector simulations","title_sort":"Normalizing flows for high-dimensional detector simulations"}],"person":[{"role":"aut","display":"Ernst, Florian","roleDisplay":"VerfasserIn","given":"Florian","family":"Ernst"},{"given":"Luigi","family":"Favaro","role":"aut","roleDisplay":"VerfasserIn","display":"Favaro, Luigi"},{"role":"aut","roleDisplay":"VerfasserIn","display":"Krause, Claudius","given":"Claudius","family":"Krause"},{"given":"Tilman","family":"Plehn","role":"aut","display":"Plehn, Tilman","roleDisplay":"VerfasserIn"},{"roleDisplay":"VerfasserIn","display":"Shih, David","role":"aut","family":"Shih","given":"David"}],"recId":"1935229141","language":["eng"],"type":{"media":"Online-Ressource","bibl":"article-journal"},"note":["Gesehen am 03.09.2025"]} 
SRT |a ERNSTFLORINORMALIZIN5202