Towards novel insights in lattice field theory with explainable machine learning
Machine learning has the potential to aid our understanding of phase structures in lattice quantum field theories through the statistical analysis of Monte Carlo samples. Available algorithms, in particular those based on deep learning, often demonstrate remarkable performance in the search for prev...
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
20 May 2020
|
| In: |
Physical review
Year: 2020, Jahrgang: 101, Heft: 9 |
| ISSN: | 2470-0029 |
| DOI: | 10.1103/PhysRevD.101.094507 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://dx.doi.org/10.1103/PhysRevD.101.094507 Verlag, lizenzpflichtig, Volltext: https://link.aps.org/doi/10.1103/PhysRevD.101.094507 
|
| Verfasserangaben: | Stefan Blücher, Lukas Kades, Jan M. Pawlowski, Nils Strodthoff and Julian M. Urban |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1702156400 | ||
| 003 | DE-627 | ||
| 005 | 20220818132931.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 200625s2020 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1103/PhysRevD.101.094507 |2 doi | |
| 035 | |a (DE-627)1702156400 | ||
| 035 | |a (DE-599)KXP1702156400 | ||
| 035 | |a (OCoLC)1341341631 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 29 |2 sdnb | ||
| 100 | 1 | |a Blücher, Stefan |d 1994- |e VerfasserIn |0 (DE-588)1178144097 |0 (DE-627)1049230787 |0 (DE-576)517694875 |4 aut | |
| 245 | 1 | 0 | |a Towards novel insights in lattice field theory with explainable machine learning |c Stefan Blücher, Lukas Kades, Jan M. Pawlowski, Nils Strodthoff and Julian M. Urban |
| 264 | 1 | |c 20 May 2020 | |
| 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 26.06.2020 | ||
| 520 | |a Machine learning has the potential to aid our understanding of phase structures in lattice quantum field theories through the statistical analysis of Monte Carlo samples. Available algorithms, in particular those based on deep learning, often demonstrate remarkable performance in the search for previously unidentified features, but tend to lack transparency if applied naively. To address these shortcomings, we propose representation learning in combination with interpretability methods as a framework for the identification of observables. More specifically, we investigate action parameter regression as a pretext task while using layer-wise relevance propagation (LRP) to identify the most important observables depending on the location in the phase diagram. The approach is put to work in the context of a scalar Yukawa model in (2+1)d. First, we investigate a multilayer perceptron to determine an importance hierarchy of several predefined, standard observables. The method is then applied directly to the raw field configurations using a convolutional network, demonstrating the ability to reconstruct all order parameters from the learned filter weights. Based on our results, we argue that due to its broad applicability, attribution methods such as LRP could prove a useful and versatile tool in our search for new physical insights. In the case of the Yukawa model, it facilitates the construction of an observable that characterizes the symmetric phase. | ||
| 700 | 1 | |a Kades, Lukas |d 1991- |e VerfasserIn |0 (DE-588)117814528X |0 (DE-627)1049231686 |0 (DE-576)517696266 |4 aut | |
| 700 | 1 | |a Pawlowski, Jan M. |d 1965- |e VerfasserIn |0 (DE-588)1047077388 |0 (DE-627)777525925 |0 (DE-576)400331381 |4 aut | |
| 700 | 1 | |a Strodthoff, Nils |d 1985- |e VerfasserIn |0 (DE-588)1032638079 |0 (DE-627)738604011 |0 (DE-576)380153076 |4 aut | |
| 700 | 1 | |a Urban, Julian M. |d 1994- |e VerfasserIn |0 (DE-588)1173444157 |0 (DE-627)1043277242 |0 (DE-576)515261769 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Physical review |d Ridge, NY : American Physical Society, 2016 |g 101(2020,9) Artikel-Nummer 094507, 12 Seiten |h Online-Ressource |w (DE-627)846313510 |w (DE-600)2844732-3 |w (DE-576)454495811 |x 2470-0029 |7 nnas |a Towards novel insights in lattice field theory with explainable machine learning |
| 773 | 1 | 8 | |g volume:101 |g year:2020 |g number:9 |a Towards novel insights in lattice field theory with explainable machine learning |
| 856 | 4 | 0 | |u https://dx.doi.org/10.1103/PhysRevD.101.094507 |x Verlag |x Resolving-System |z lizenzpflichtig |3 Volltext |
| 856 | 4 | 0 | |u https://link.aps.org/doi/10.1103/PhysRevD.101.094507 |x Verlag |z lizenzpflichtig |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20200625 | ||
| 993 | |a Article | ||
| 994 | |a 2020 | ||
| 998 | |g 1173444157 |a Urban, Julian M. |m 1173444157:Urban, Julian M. |d 130000 |d 700000 |d 728500 |e 130000PU1173444157 |e 700000PU1173444157 |e 728500PU1173444157 |k 0/130000/ |k 0/700000/ |k 1/700000/728500/ |p 5 |y j | ||
| 998 | |g 1032638079 |a Strodthoff, Nils |m 1032638079:Strodthoff, Nils |p 4 | ||
| 998 | |g 1047077388 |a Pawlowski, Jan M. |m 1047077388:Pawlowski, Jan M. |d 130000 |d 130300 |d 700000 |d 728500 |e 130000PP1047077388 |e 130300PP1047077388 |e 700000PP1047077388 |e 728500PP1047077388 |k 0/130000/ |k 1/130000/130300/ |k 0/700000/ |k 1/700000/728500/ |p 3 | ||
| 998 | |g 117814528X |a Kades, Lukas |m 117814528X:Kades, Lukas |d 130000 |d 130300 |d 700000 |d 728500 |e 130000PK117814528X |e 130300PK117814528X |e 700000PK117814528X |e 728500PK117814528X |k 0/130000/ |k 1/130000/130300/ |k 0/700000/ |k 1/700000/728500/ |p 2 | ||
| 998 | |g 1178144097 |a Blücher, Stefan |m 1178144097:Blücher, Stefan |p 1 |x j | ||
| 999 | |a KXP-PPN1702156400 |e 3691042589 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"recId":"1702156400","language":["eng"],"name":{"displayForm":["Stefan Blücher, Lukas Kades, Jan M. Pawlowski, Nils Strodthoff and Julian M. Urban"]},"type":{"bibl":"article-journal","media":"Online-Ressource"},"id":{"eki":["1702156400"],"doi":["10.1103/PhysRevD.101.094507"]},"origin":[{"dateIssuedDisp":"20 May 2020","dateIssuedKey":"2020"}],"note":["Gesehen am 26.06.2020"],"relHost":[{"title":[{"title":"Physical review","title_sort":"Physical review"}],"part":{"text":"101(2020,9) Artikel-Nummer 094507, 12 Seiten","issue":"9","year":"2020","volume":"101"},"corporate":[{"display":"American Physical Society","role":"isb"}],"disp":"Towards novel insights in lattice field theory with explainable machine learningPhysical review","type":{"media":"Online-Ressource","bibl":"periodical"},"language":["eng"],"titleAlt":[{"title":"Particles, fields, gravitation, and cosmology"}],"physDesc":[{"extent":"Online-Ressource"}],"note":["Gesehen am 14.03.2023"],"origin":[{"dateIssuedDisp":"2016-","publisher":"American Physical Society","dateIssuedKey":"2016","publisherPlace":"Ridge, NY"}],"pubHistory":["3rd series, volume 93, number 1 (January 2016)-"],"id":{"zdb":["2844732-3"],"eki":["846313510"],"issn":["2470-0029"]},"recId":"846313510","name":{"displayForm":["published by American Physical Society"]}}],"title":[{"title_sort":"Towards novel insights in lattice field theory with explainable machine learning","title":"Towards novel insights in lattice field theory with explainable machine learning"}],"person":[{"family":"Blücher","given":"Stefan","display":"Blücher, Stefan","role":"aut"},{"given":"Lukas","family":"Kades","display":"Kades, Lukas","role":"aut"},{"display":"Pawlowski, Jan M.","role":"aut","family":"Pawlowski","given":"Jan M."},{"given":"Nils","family":"Strodthoff","display":"Strodthoff, Nils","role":"aut"},{"given":"Julian M.","family":"Urban","display":"Urban, Julian M.","role":"aut"}]} | ||
| SRT | |a BLUECHERSTTOWARDSNOV2020 | ||