View in EDS

Neural network analysis of neutron and X‐ray reflectivity data incorporating prior knowledge.

Saved in:
Bibliographic Details
Title: Neural network analysis of neutron and X‐ray reflectivity data incorporating prior knowledge.
Authors: Munteanu, Valentin1 (AUTHOR) valentin.munteanu@uni-tuebingen.de, Starostin, Vladimir1 (AUTHOR), Greco, Alessandro1 (AUTHOR), Pithan, Linus1,2 (AUTHOR), Gerlach, Alexander1 (AUTHOR), Hinderhofer, Alexander1 (AUTHOR), Kowarik, Stefan3 (AUTHOR), Schreiber, Frank1 (AUTHOR) frank.schreiber@uni-tuebingen.de
Source: Journal of Applied Crystallography. Apr2024, Vol. 57 Issue 2, p456-469. 14p.
Subjects: Neutron reflectivity, Scattering (Physics), Prior learning, Inverse problems, Thin films
Abstract: Due to the ambiguity related to the lack of phase information, determining the physical parameters of multilayer thin films from measured neutron and X‐ray reflectivity curves is, on a fundamental level, an underdetermined inverse problem. This ambiguity poses limitations on standard neural networks, constraining the range and number of considered parameters in previous machine learning solutions. To overcome this challenge, a novel training procedure has been designed which incorporates dynamic prior boundaries for each physical parameter as additional inputs to the neural network. In this manner, the neural network can be trained simultaneously on all well‐posed subintervals of a larger parameter space in which the inverse problem is underdetermined. During inference, users can flexibly input their own prior knowledge about the physical system to constrain the neural network prediction to distinct target subintervals in the parameter space. The effectiveness of the method is demonstrated in various scenarios, including multilayer structures with a box model parameterization and a physics‐inspired special parameterization of the scattering length density profile for a multilayer structure. In contrast to previous methods, this approach scales favourably when increasing the complexity of the inverse problem, working properly even for a five‐layer multilayer model and a periodic multilayer model with up to 17 open parameters. [ABSTRACT FROM AUTHOR]
Copyright of Journal of Applied Crystallography is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 176496410
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Neural network analysis of neutron and X‐ray reflectivity data incorporating prior knowledge.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Munteanu%2C+Valentin%22">Munteanu, Valentin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> valentin.munteanu@uni-tuebingen.de</i><br /><searchLink fieldCode="AR" term="%22Starostin%2C+Vladimir%22">Starostin, Vladimir</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Greco%2C+Alessandro%22">Greco, Alessandro</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Pithan%2C+Linus%22">Pithan, Linus</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Gerlach%2C+Alexander%22">Gerlach, Alexander</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hinderhofer%2C+Alexander%22">Hinderhofer, Alexander</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Kowarik%2C+Stefan%22">Kowarik, Stefan</searchLink><relatesTo>3</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Schreiber%2C+Frank%22">Schreiber, Frank</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> frank.schreiber@uni-tuebingen.de</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Journal+of+Applied+Crystallography%22">Journal of Applied Crystallography</searchLink>. Apr2024, Vol. 57 Issue 2, p456-469. 14p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Neutron+reflectivity%22">Neutron reflectivity</searchLink><br /><searchLink fieldCode="DE" term="%22Scattering+%28Physics%29%22">Scattering (Physics)</searchLink><br /><searchLink fieldCode="DE" term="%22Prior+learning%22">Prior learning</searchLink><br /><searchLink fieldCode="DE" term="%22Inverse+problems%22">Inverse problems</searchLink><br /><searchLink fieldCode="DE" term="%22Thin+films%22">Thin films</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Due to the ambiguity related to the lack of phase information, determining the physical parameters of multilayer thin films from measured neutron and X‐ray reflectivity curves is, on a fundamental level, an underdetermined inverse problem. This ambiguity poses limitations on standard neural networks, constraining the range and number of considered parameters in previous machine learning solutions. To overcome this challenge, a novel training procedure has been designed which incorporates dynamic prior boundaries for each physical parameter as additional inputs to the neural network. In this manner, the neural network can be trained simultaneously on all well‐posed subintervals of a larger parameter space in which the inverse problem is underdetermined. During inference, users can flexibly input their own prior knowledge about the physical system to constrain the neural network prediction to distinct target subintervals in the parameter space. The effectiveness of the method is demonstrated in various scenarios, including multilayer structures with a box model parameterization and a physics‐inspired special parameterization of the scattering length density profile for a multilayer structure. In contrast to previous methods, this approach scales favourably when increasing the complexity of the inverse problem, working properly even for a five‐layer multilayer model and a periodic multilayer model with up to 17 open parameters. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Journal of Applied Crystallography is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=176496410
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1107/S1600576724002115
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 14
        StartPage: 456
    Subjects:
      – SubjectFull: Neutron reflectivity
        Type: general
      – SubjectFull: Scattering (Physics)
        Type: general
      – SubjectFull: Prior learning
        Type: general
      – SubjectFull: Inverse problems
        Type: general
      – SubjectFull: Thin films
        Type: general
    Titles:
      – TitleFull: Neural network analysis of neutron and X‐ray reflectivity data incorporating prior knowledge.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Munteanu, Valentin
      – PersonEntity:
          Name:
            NameFull: Starostin, Vladimir
      – PersonEntity:
          Name:
            NameFull: Greco, Alessandro
      – PersonEntity:
          Name:
            NameFull: Pithan, Linus
      – PersonEntity:
          Name:
            NameFull: Gerlach, Alexander
      – PersonEntity:
          Name:
            NameFull: Hinderhofer, Alexander
      – PersonEntity:
          Name:
            NameFull: Kowarik, Stefan
      – PersonEntity:
          Name:
            NameFull: Schreiber, Frank
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 04
              Text: Apr2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 00218898
          Numbering:
            – Type: volume
              Value: 57
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Journal of Applied Crystallography
              Type: main
ResultId 1