Report
Προβολή σε EDS

Benchmarking Global Optimizers.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Benchmarking Global Optimizers.
Συγγραφείς: Arnoud, Antoine, Guvenen, Fatih, Kleineberg, Tatjana
Πηγή: Working Papers Series (Federal Reserve Bank of Minneapolis); Dec2023, Issue 801, following p1-58, 59p
Θεματικοί όροι: Optimization algorithms, Benchmarking (Management), Panel analysis, Calibration, Parallelizing compilers
Περίληψη: Benchmarking Global Optimizers* Antoine Arnoud? Fatih Guvenen? Tatjana Kleineberg§ October 19, 2023 Abstract We benchmark six global optimization algorithms by comparing their performance on challenging multidimensional test functions as well as on a method of simulated moments estimation of a panel data model of earnings dynamics. Five of the algorithms are from the popular NLopt open-source library: (i) Controlled Random Search with local mutation (CRS), (ii) Improved Stochastic Ranking Evolution Strategy (ISRES), (iii) Multi-Level Single-Linkage (MLSL), (iv) Stochastic Global Optimization (StoGo), and (v) Evolutionary Strategy with Cauchy distribution (ESCH). The sixth algorithm is TikTak, which is a multistart global optimization algorithm used in some recent economic applications. For completeness, we add three popular local algorithms to the comparison--the Nelder-Mead downhill simplex algorithm, the Derivative-Free Nonlinear Least Squares (DFNLS) algorithm, and a popular variant of the Davidon-Fletcher-Powell (DFPMIN) algorithm. To give a detailed comparison of algorithms, we use benchmarking tools recently developed in the optimization literature. We find that the success rate of many optimizers varies dramatically with the characteristics of each problem and the computational budget that is available. Overall, TikTak is the strongest performer both on the test functions and the economic application. The next-best performing optimizers are StoGo for the test functions and MLSL and ISRES for the economic application. [ABSTRACT FROM AUTHOR]
Copyright of Working Papers Series (Federal Reserve Bank of Minneapolis) is the property of Federal Reserve Bank of Minneapolis 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.)
Βάση Δεδομένων: Complementary Index
FullText Text:
  Availability: 0
Header DbId: edb
DbLabel: Complementary Index
An: 174431088
RelevancyScore: 965
AccessLevel: 6
PubType: Report
PubTypeId: report
PreciseRelevancyScore: 964.691833496094
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Benchmarking Global Optimizers.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Arnoud%2C+Antoine%22">Arnoud, Antoine</searchLink><br /><searchLink fieldCode="AR" term="%22Guvenen%2C+Fatih%22">Guvenen, Fatih</searchLink><br /><searchLink fieldCode="AR" term="%22Kleineberg%2C+Tatjana%22">Kleineberg, Tatjana</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Working Papers Series (Federal Reserve Bank of Minneapolis); Dec2023, Issue 801, following p1-58, 59p
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Optimization+algorithms%22">Optimization algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Benchmarking+%28Management%29%22">Benchmarking (Management)</searchLink><br /><searchLink fieldCode="DE" term="%22Panel+analysis%22">Panel analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Calibration%22">Calibration</searchLink><br /><searchLink fieldCode="DE" term="%22Parallelizing+compilers%22">Parallelizing compilers</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Benchmarking Global Optimizers* Antoine Arnoud? Fatih Guvenen? Tatjana Kleineberg§ October 19, 2023 Abstract We benchmark six global optimization algorithms by comparing their performance on challenging multidimensional test functions as well as on a method of simulated moments estimation of a panel data model of earnings dynamics. Five of the algorithms are from the popular NLopt open-source library: (i) Controlled Random Search with local mutation (CRS), (ii) Improved Stochastic Ranking Evolution Strategy (ISRES), (iii) Multi-Level Single-Linkage (MLSL), (iv) Stochastic Global Optimization (StoGo), and (v) Evolutionary Strategy with Cauchy distribution (ESCH). The sixth algorithm is TikTak, which is a multistart global optimization algorithm used in some recent economic applications. For completeness, we add three popular local algorithms to the comparison--the Nelder-Mead downhill simplex algorithm, the Derivative-Free Nonlinear Least Squares (DFNLS) algorithm, and a popular variant of the Davidon-Fletcher-Powell (DFPMIN) algorithm. To give a detailed comparison of algorithms, we use benchmarking tools recently developed in the optimization literature. We find that the success rate of many optimizers varies dramatically with the characteristics of each problem and the computational budget that is available. Overall, TikTak is the strongest performer both on the test functions and the economic application. The next-best performing optimizers are StoGo for the test functions and MLSL and ISRES for the economic application. [ABSTRACT FROM AUTHOR]
– Name: Abstract
  Label:
  Group: Ab
  Data: <i>Copyright of Working Papers Series (Federal Reserve Bank of Minneapolis) is the property of Federal Reserve Bank of Minneapolis 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=edb&AN=174431088
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.21034/wp.801
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 59
        StartPage: 1
    Subjects:
      – SubjectFull: Optimization algorithms
        Type: general
      – SubjectFull: Benchmarking (Management)
        Type: general
      – SubjectFull: Panel analysis
        Type: general
      – SubjectFull: Calibration
        Type: general
      – SubjectFull: Parallelizing compilers
        Type: general
    Titles:
      – TitleFull: Benchmarking Global Optimizers.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Arnoud, Antoine
      – PersonEntity:
          Name:
            NameFull: Guvenen, Fatih
      – PersonEntity:
          Name:
            NameFull: Kleineberg, Tatjana
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 12
              Text: Dec2023
              Type: published
              Y: 2023
          Numbering:
            – Type: issue
              Value: 801
          Titles:
            – TitleFull: Working Papers Series (Federal Reserve Bank of Minneapolis)
              Type: main
ResultId 1