Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
We examine unsupervised machine learning techniques to learn features that best describe configurations of the two-dimensional Ising model and the three-dimensional XY model. The methods range from principal component analysis over manifold and clustering methods to artificial neural-network-based v...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
18 August 2017
|
| In: |
Physical review
Year: 2017, Volume: 96, Issue: 2, Pages: 022140 |
| ISSN: | 2470-0053 |
| DOI: | 10.1103/PhysRevE.96.022140 |
| Online Access: | Verlag, Volltext: http://dx.doi.org/10.1103/PhysRevE.96.022140 Verlag, Volltext: https://link.aps.org/doi/10.1103/PhysRevE.96.022140 
|
| Author Notes: | Sebastian J. Wetzel |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1572234946 | ||
| 003 | DE-627 | ||
| 005 | 20220814121424.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 180423s2017 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1103/PhysRevE.96.022140 |2 doi | |
| 035 | |a (DE-627)1572234946 | ||
| 035 | |a (DE-576)502234946 | ||
| 035 | |a (DE-599)BSZ502234946 | ||
| 035 | |a (OCoLC)1341007519 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 29 |2 sdnb | ||
| 100 | 1 | |a Wetzel, Sebastian |d 1987- |e VerfasserIn |0 (DE-588)1085180069 |0 (DE-627)848860608 |0 (DE-576)456916016 |4 aut | |
| 245 | 1 | 0 | |a Unsupervised learning of phase transitions |b from principal component analysis to variational autoencoders |c Sebastian J. Wetzel |
| 264 | 1 | |c 18 August 2017 | |
| 300 | |a 11 | ||
| 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 23.04.2018 | ||
| 520 | |a We examine unsupervised machine learning techniques to learn features that best describe configurations of the two-dimensional Ising model and the three-dimensional XY model. The methods range from principal component analysis over manifold and clustering methods to artificial neural-network-based variational autoencoders. They are applied to Monte Carlo-sampled configurations and have, a priori, no knowledge about the Hamiltonian or the order parameter. We find that the most promising algorithms are principal component analysis and variational autoencoders. Their predicted latent parameters correspond to the known order parameters. The latent representations of the models in question are clustered, which makes it possible to identify phases without prior knowledge of their existence. Furthermore, we find that the reconstruction loss function can be used as a universal identifier for phase transitions. | ||
| 773 | 0 | 8 | |i Enthalten in |t Physical review |d Woodbury, NY : Inst., 2016 |g 96(2017), 2, Seite 022140 |h Online-Ressource |w (DE-627)846123010 |w (DE-600)2844562-4 |w (DE-576)454423063 |x 2470-0053 |7 nnas |a Unsupervised learning of phase transitions from principal component analysis to variational autoencoders |
| 773 | 1 | 8 | |g volume:96 |g year:2017 |g number:2 |g pages:022140 |g extent:11 |a Unsupervised learning of phase transitions from principal component analysis to variational autoencoders |
| 856 | 4 | 0 | |u http://dx.doi.org/10.1103/PhysRevE.96.022140 |x Verlag |x Resolving-System |3 Volltext |
| 856 | 4 | 0 | |u https://link.aps.org/doi/10.1103/PhysRevE.96.022140 |x Verlag |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20180423 | ||
| 993 | |a Article | ||
| 994 | |a 2017 | ||
| 998 | |g 1085180069 |a Wetzel, Sebastian |m 1085180069:Wetzel, Sebastian |d 130000 |d 130300 |e 130000PW1085180069 |e 130300PW1085180069 |k 0/130000/ |k 1/130000/130300/ |p 1 |x j |y j | ||
| 999 | |a KXP-PPN1572234946 |e 3006751248 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"origin":[{"dateIssuedKey":"2017","dateIssuedDisp":"18 August 2017"}],"id":{"doi":["10.1103/PhysRevE.96.022140"],"eki":["1572234946"]},"name":{"displayForm":["Sebastian J. Wetzel"]},"physDesc":[{"extent":"11 S."}],"relHost":[{"title":[{"title_sort":"Physical review","title":"Physical review"}],"part":{"year":"2017","issue":"2","pages":"022140","text":"96(2017), 2, Seite 022140","volume":"96","extent":"11"},"titleAlt":[{"title":"Statistical, nonlinear, and soft matter physics"}],"pubHistory":["Vol. 93, Iss. 1, January 2016-"],"recId":"846123010","language":["eng"],"corporate":[{"roleDisplay":"Herausgebendes Organ","display":"American Institute of Physics","role":"isb"},{"role":"isb","roleDisplay":"Herausgebendes Organ","display":"American Physical Society"}],"type":{"bibl":"periodical","media":"Online-Ressource"},"disp":"Unsupervised learning of phase transitions from principal component analysis to variational autoencodersPhysical review","id":{"issn":["2470-0053"],"zdb":["2844562-4"],"eki":["846123010"]},"origin":[{"dateIssuedDisp":"January 2016-","publisher":"Inst.","publisherPlace":"Woodbury, NY"}],"name":{"displayForm":["publ. by The American Institute of Physics"]},"physDesc":[{"extent":"Online-Ressource"}]}],"title":[{"subtitle":"from principal component analysis to variational autoencoders","title":"Unsupervised learning of phase transitions","title_sort":"Unsupervised learning of phase transitions"}],"person":[{"given":"Sebastian","family":"Wetzel","role":"aut","roleDisplay":"VerfasserIn","display":"Wetzel, Sebastian"}],"note":["Gesehen am 23.04.2018"],"type":{"bibl":"article-journal","media":"Online-Ressource"},"language":["eng"],"recId":"1572234946"} | ||
| SRT | |a WETZELSEBAUNSUPERVIS1820 | ||