Exoplanet characterization using conditional invertible neural networks
Context. The characterization of the interior of an exoplanet is an inverse problem. The solution requires statistical methods such as Bayesian inference. Current methods employ Markov chain Monte Carlo (MCMC) sampling to infer the posterior probability of the planetary structure parameters for a gi...
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
20 April 2023
|
| In: |
Astronomy and astrophysics
Year: 2023, Jahrgang: 672, Pages: 1-16 |
| ISSN: | 1432-0746 |
| DOI: | 10.1051/0004-6361/202243230 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.1051/0004-6361/202243230 Verlag, kostenfrei, Volltext: https://www.aanda.org/articles/aa/abs/2023/04/aa43230-22/aa43230-22.html 
|
| Verfasserangaben: | Jonas Haldemann, Victor Ksoll, Daniel Walter, Yann Alibert, Ralf S. Klessen, Willy Benz, Ullrich Koethe, Lynton Ardizzone, and Carsten Rother |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1851263799 | ||
| 003 | DE-627 | ||
| 005 | 20230706201634.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 230629s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1051/0004-6361/202243230 |2 doi | |
| 035 | |a (DE-627)1851263799 | ||
| 035 | |a (DE-599)KXP1851263799 | ||
| 035 | |a (OCoLC)1389527662 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 29 |2 sdnb | ||
| 100 | 1 | |a Haldemann, Jonas |e VerfasserIn |0 (DE-588)1270157841 |0 (DE-627)181885483X |4 aut | |
| 245 | 1 | 0 | |a Exoplanet characterization using conditional invertible neural networks |c Jonas Haldemann, Victor Ksoll, Daniel Walter, Yann Alibert, Ralf S. Klessen, Willy Benz, Ullrich Koethe, Lynton Ardizzone, and Carsten Rother |
| 264 | 1 | |c 20 April 2023 | |
| 300 | |a 16 | ||
| 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 29.06.2023 | ||
| 520 | |a Context. The characterization of the interior of an exoplanet is an inverse problem. The solution requires statistical methods such as Bayesian inference. Current methods employ Markov chain Monte Carlo (MCMC) sampling to infer the posterior probability of the planetary structure parameters for a given exoplanet. These methods are time-consuming because they require the evaluation of a planetary structure model ~105 times. Aims. To speed up the inference process when characterizing an exoplanet, we propose to use conditional invertible neural networks to calculate the posterior probability of the planetary structure parameters.Methods. Conditional invertible neural networks (cINNs) are a special type of neural network that excels at solving inverse problems. We constructed a cINN following the framework for easily invertible architectures (FreIA). This neural network was then trained on a database of 5.6 × 106 internal structure models to recover the inverse mapping between internal structure parameters and observable features (i.e., planetary mass, planetary radius, and elemental composition of the host star). We also show how observational uncertainties can be accounted for. Results. The cINN method was compared to a commonly used Metropolis-Hastings MCMC. To do this, we repeated the characterization of the exoplanet K2-111 b, using both the MCMC method and the trained cINN. We show that the inferred posterior probability distributions of the internal structure parameters from both methods are very similar; the largest differences are seen in the exoplanet water content. Thus, cINNs are a possible alternative to the standard time-consuming sampling methods. cINNs allow infering the composition of an exoplanet that is orders of magnitude faster than what is possible using an MCMC method. The computation of a large database of internal structures to train the neural network is still required, however. Because this database is only computed once, we found that using an invertible neural network is more efficient than an MCMC when more than ten exoplanets are characterized using the same neural network. | ||
| 700 | 1 | |a Ksoll, Victor F. |d 1992- |e VerfasserIn |0 (DE-588)1176827448 |0 (DE-627)1048198030 |0 (DE-576)51686372X |4 aut | |
| 700 | 1 | |a Walter, Daniel |e VerfasserIn |0 (DE-588)1270157574 |0 (DE-627)1818854759 |4 aut | |
| 700 | 1 | |a Alibert, Yann |e VerfasserIn |4 aut | |
| 700 | 1 | |a Klessen, Ralf S. |d 1968- |e VerfasserIn |0 (DE-588)120533820 |0 (DE-627)392381532 |0 (DE-576)178685399 |4 aut | |
| 700 | 1 | |a Benz, Willy |e VerfasserIn |4 aut | |
| 700 | 1 | |a Köthe, Ullrich |e VerfasserIn |0 (DE-588)123963435 |0 (DE-627)594480884 |0 (DE-576)304484520 |4 aut | |
| 700 | 1 | |a Ardizzone, Lynton |d 1994- |e VerfasserIn |0 (DE-588)1194988512 |0 (DE-627)1677182296 |4 aut | |
| 700 | 1 | |a Rother, Carsten |e VerfasserIn |0 (DE-588)1181464692 |0 (DE-627)1662676883 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Astronomy and astrophysics |d Les Ulis : EDP Sciences, 1969 |g 672(2023) vom: Apr., Artikel-ID A180, Seite 1-16 |h Online-Ressource |w (DE-627)253390222 |w (DE-600)1458466-9 |w (DE-576)072283351 |x 1432-0746 |7 nnas |a Exoplanet characterization using conditional invertible neural networks |
| 773 | 1 | 8 | |g volume:672 |g year:2023 |g month:04 |g elocationid:A180 |g pages:1-16 |g extent:16 |a Exoplanet characterization using conditional invertible neural networks |
| 856 | 4 | 0 | |u https://doi.org/10.1051/0004-6361/202243230 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u https://www.aanda.org/articles/aa/abs/2023/04/aa43230-22/aa43230-22.html |x Verlag |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20230629 | ||
| 993 | |a Article | ||
| 994 | |a 2023 | ||
| 998 | |g 1181464692 |a Rother, Carsten |m 1181464692:Rother, Carsten |d 700000 |d 708070 |e 700000PR1181464692 |e 708070PR1181464692 |k 0/700000/ |k 1/700000/708070/ |p 9 |y j | ||
| 998 | |g 1194988512 |a Ardizzone, Lynton |m 1194988512:Ardizzone, Lynton |d 110000 |e 110000PA1194988512 |k 0/110000/ |p 8 | ||
| 998 | |g 123963435 |a Köthe, Ullrich |m 123963435:Köthe, Ullrich |d 700000 |d 708070 |e 700000PK123963435 |e 708070PK123963435 |k 0/700000/ |k 1/700000/708070/ |p 7 | ||
| 998 | |g 120533820 |a Klessen, Ralf S. |m 120533820:Klessen, Ralf S. |d 700000 |d 714000 |d 714200 |e 700000PK120533820 |e 714000PK120533820 |e 714200PK120533820 |k 0/700000/ |k 1/700000/714000/ |k 2/700000/714000/714200/ |p 5 | ||
| 998 | |g 1270157574 |a Walter, Daniel |m 1270157574:Walter, Daniel |p 3 | ||
| 998 | |g 1176827448 |a Ksoll, Victor F. |m 1176827448:Ksoll, Victor F. |d 700000 |d 714000 |d 714200 |e 700000PK1176827448 |e 714000PK1176827448 |e 714200PK1176827448 |k 0/700000/ |k 1/700000/714000/ |k 2/700000/714000/714200/ |p 2 | ||
| 999 | |a KXP-PPN1851263799 |e 4344652231 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"person":[{"display":"Haldemann, Jonas","given":"Jonas","family":"Haldemann","role":"aut"},{"given":"Victor F.","display":"Ksoll, Victor F.","family":"Ksoll","role":"aut"},{"family":"Walter","role":"aut","given":"Daniel","display":"Walter, Daniel"},{"display":"Alibert, Yann","given":"Yann","role":"aut","family":"Alibert"},{"given":"Ralf S.","display":"Klessen, Ralf S.","family":"Klessen","role":"aut"},{"family":"Benz","role":"aut","given":"Willy","display":"Benz, Willy"},{"given":"Ullrich","display":"Köthe, Ullrich","role":"aut","family":"Köthe"},{"family":"Ardizzone","role":"aut","given":"Lynton","display":"Ardizzone, Lynton"},{"given":"Carsten","display":"Rother, Carsten","role":"aut","family":"Rother"}],"recId":"1851263799","note":["Gesehen am 29.06.2023"],"language":["eng"],"name":{"displayForm":["Jonas Haldemann, Victor Ksoll, Daniel Walter, Yann Alibert, Ralf S. Klessen, Willy Benz, Ullrich Koethe, Lynton Ardizzone, and Carsten Rother"]},"relHost":[{"language":["eng"],"part":{"extent":"16","pages":"1-16","text":"672(2023) vom: Apr., Artikel-ID A180, Seite 1-16","year":"2023","volume":"672"},"pubHistory":["1.1969 -"],"note":["Gesehen am 21.06.2024","Erscheint 36mal jährlich in 12 Bänden zu je 3 Ausgaben","Fortsetzung der Druck-Ausgabe"],"corporate":[{"role":"isb","display":"European Southern Observatory"}],"disp":"Exoplanet characterization using conditional invertible neural networksAstronomy and astrophysics","type":{"bibl":"periodical","media":"Online-Ressource"},"titleAlt":[{"title":"Astronomy & astrophysics"},{"title":"a European journal"}],"name":{"displayForm":["European Southern Observatory (ESO)"]},"recId":"253390222","title":[{"title_sort":"Astronomy and astrophysics","title":"Astronomy and astrophysics","subtitle":"an international weekly journal"}],"id":{"eki":["253390222"],"issn":["1432-0746"],"zdb":["1458466-9"]},"physDesc":[{"extent":"Online-Ressource"}],"origin":[{"dateIssuedDisp":"1969-","publisher":"EDP Sciences ; Springer","publisherPlace":"Les Ulis ; Berlin ; Heidelberg","dateIssuedKey":"1969"}]}],"origin":[{"dateIssuedDisp":"20 April 2023","dateIssuedKey":"2023"}],"title":[{"title":"Exoplanet characterization using conditional invertible neural networks","title_sort":"Exoplanet characterization using conditional invertible neural networks"}],"type":{"bibl":"article-journal","media":"Online-Ressource"},"id":{"eki":["1851263799"],"doi":["10.1051/0004-6361/202243230"]},"physDesc":[{"extent":"16 S."}]} | ||
| SRT | |a HALDEMANNJEXOPLANETC2020 | ||