IDEAS home Printed from https://ideas.repec.org/a/spr/compst/v34y2019i3d10.1007_s00180-018-0832-9.html
   My bibliography  Save this article

Sparse kernel deep stacking networks

Author

Listed:
  • Thomas Welchowski

    (University Hospital Bonn)

  • Matthias Schmid

    (University Hospital Bonn)

Abstract

This article introduces the supervised deep learning method sparse kernel deep stacking networks (SKDSNs), which extend traditional kernel deep stacking networks (KDSNs) by incorporating a set of data-driven regularization and variable selection steps to improve predictive performance in high-dimensional settings. Before model fitting, variable pre-selection is carried out using genetic algorithms in combination with the randomized dependence coefficient, accounting for non-linear dependencies among the inputs and the outcome variable. During model fitting, internal variable selection is based on a ranked feature ordering which is tuned within the model-based optimization framework. Further regularization steps include $$L_1$$ L 1 -penalized kernel regression and dropout. Our simulation studies demonstrate an improved prediction accuracy of SKDSNs compared to traditional KDSNs. Runtime analysis of SKDSNs shows that the dimension of the random Fourier transformation greatly affects computational efficiency, and that the speed of SKDSNs can be improved by applying a subsampling-based ensemble strategy. Numerical experiments show that the latter strategy further increases predictive performance. Application of SKDSNs to three biomedical data sets confirm the results of the simulation study. SKDSNs are implemented in a new version of the R package kernDeepStackNet.

Suggested Citation

  • Thomas Welchowski & Matthias Schmid, 2019. "Sparse kernel deep stacking networks," Computational Statistics, Springer, vol. 34(3), pages 993-1014, September.
    • Handle: RePEc:spr:compst:v:34:y:2019:i:3:d:10.1007_s00180-018-0832-9
    DOI: 10.1007/s00180-018-0832-9
    as

    Download full text from publisher

    File URL: http://link.springer.com/10.1007/s00180-018-0832-9
    File Function: Abstract
    Download Restriction: Access to the full text of the articles in this series is restricted.
    File URL: https://libkey.io/10.1007/s00180-018-0832-9?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    References listed on IDEAS

    as
    1. Benjamin Hofner & Andreas Mayr & Nikolay Robinzonov & Matthias Schmid, 2014. "Model-based boosting in R: a hands-on tutorial using the R package mboost," Computational Statistics, Springer, vol. 29(1), pages 3-35, February.
    2. Roustant, Olivier & Ginsbourger, David & Deville, Yves, 2012. "DiceKriging, DiceOptim: Two R Packages for the Analysis of Computer Experiments by Kriging-Based Metamodeling and Optimization," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 51(i01).
    3. Scrucca, Luca, 2013. "GA: A Package for Genetic Algorithms in R," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 53(i04).
    4. López-Ibáñez, Manuel & Dubois-Lacoste, Jérémie & Pérez Cáceres, Leslie & Birattari, Mauro & Stützle, Thomas, 2016. "The irace package: Iterated racing for automatic algorithm configuration," Operations Research Perspectives, Elsevier, vol. 3(C), pages 43-58.
    5. Tsallis, Constantino & Stariolo, Daniel A., 1996. "Generalized simulated annealing," Physica A: Statistical Mechanics and its Applications, Elsevier, vol. 233(1), pages 395-406.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Matthias Schmid & Bernd Bischl & Hans A. Kestler, 2019. "Proceedings of Reisensburg 2016–2017," Computational Statistics, Springer, vol. 34(3), pages 943-944, September.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Olgun Aydin & Bartłomiej Igliński & Krzysztof Krukowski & Marek Siemiński, 2022. "Analyzing Wind Energy Potential Using Efficient Global Optimization: A Case Study for the City Gdańsk in Poland," Energies, MDPI, vol. 15(9), pages 1-22, April.
    2. Krityakierne, Tipaluck & Baowan, Duangkamon, 2020. "Aggregated GP-based Optimization for Contaminant Source Localization," Operations Research Perspectives, Elsevier, vol. 7(C).
    3. Ehsan Mehdad & Jack P. C. Kleijnen, 2018. "Efficient global optimisation for black-box simulation via sequential intrinsic Kriging," Journal of the Operational Research Society, Taylor & Francis Journals, vol. 69(11), pages 1725-1737, November.
    4. Bergeaud, Antonin & Raimbault, Juste, 2020. "An empirical analysis of the spatial variability of fuel prices in the United States," Transportation Research Part A: Policy and Practice, Elsevier, vol. 132(C), pages 131-143.
    5. Asghari, Mohammad & Jaber, Mohamad Y. & Mirzapour Al-e-hashem, S.M.J., 2023. "Coordinating vessel recovery actions: Analysis of disruption management in a liner shipping service," European Journal of Operational Research, Elsevier, vol. 307(2), pages 627-644.
    6. Alex Gliesch & Marcus Ritt, 2022. "A new heuristic for finding verifiable k-vertex-critical subgraphs," Journal of Heuristics, Springer, vol. 28(1), pages 61-91, February.
    7. Carolina G. Marcelino & João V. C. Avancini & Carla A. D. M. Delgado & Elizabeth F. Wanner & Silvia Jiménez-Fernández & Sancho Salcedo-Sanz, 2021. "Dynamic Electric Dispatch for Wind Power Plants: A New Automatic Controller System Using Evolutionary Algorithms," Sustainability, MDPI, vol. 13(21), pages 1-20, October.
    8. Rubenthaler, Sylvain & Rydén, Tobias & Wiktorsson, Magnus, 2009. "Fast simulated annealing in with an application to maximum likelihood estimation in state-space models," Stochastic Processes and their Applications, Elsevier, vol. 119(6), pages 1912-1931, June.
    9. Riccardo De Bin & Vegard Grødem Stikbakke, 2023. "A boosting first-hitting-time model for survival analysis in high-dimensional settings," Lifetime Data Analysis: An International Journal Devoted to Statistical Methods and Applications for Time-to-Event Data, Springer, vol. 29(2), pages 420-440, April.
    10. Diariétou Sambakhé & Lauriane Rouan & Jean-Noël Bacro & Eric Gozé, 2019. "Conditional optimization of a noisy function using a kriging metamodel," Journal of Global Optimization, Springer, vol. 73(3), pages 615-636, March.
    11. Lazzari, Florencia & Mor, Gerard & Cipriano, Jordi & Solsona, Francesc & Chemisana, Daniel & Guericke, Daniela, 2023. "Optimizing planning and operation of renewable energy communities with genetic algorithms," Applied Energy, Elsevier, vol. 338(C).
    12. Biewen, Martin & Kugler, Philipp, 2021. "Two-stage least squares random forests with an application to Angrist and Evans (1998)," Economics Letters, Elsevier, vol. 204(C).
    13. Xuefei Lu & Alessandro Rudi & Emanuele Borgonovo & Lorenzo Rosasco, 2020. "Faster Kriging: Facing High-Dimensional Simulators," Operations Research, INFORMS, vol. 68(1), pages 233-249, January.
    14. Moriguchi, Kai & Ueki, Tatsuhito & Saito, Masashi, 2020. "Establishing optimal forest harvesting regulation with continuous approximation," Operations Research Perspectives, Elsevier, vol. 7(C).
    15. Castellares, Fredy & Patrício, Silvio C. & Lemonte, Artur J. & Queiroz, Bernardo L., 2020. "On closed-form expressions to Gompertz–Makeham life expectancy," Theoretical Population Biology, Elsevier, vol. 134(C), pages 53-60.
    16. Dirick, Lore & Claeskens, Gerda & Baesens, Bart, 2015. "An Akaike information criterion for multiple event mixture cure models," European Journal of Operational Research, Elsevier, vol. 241(2), pages 449-457.
    17. Véronique François & Yasemin Arda & Yves Crama, 2019. "Adaptive Large Neighborhood Search for Multitrip Vehicle Routing with Time Windows," Transportation Science, INFORMS, vol. 53(6), pages 1706-1730, November.
    18. Chang-Yong Lee & Dongju Lee, 2014. "Determination of initial temperature in fast simulated annealing," Computational Optimization and Applications, Springer, vol. 58(2), pages 503-522, June.
    19. Ofer M. Shir & Xi. Xing & Herschel. Rabitz, 2021. "Multi-level evolution strategies for high-resolution black-box control," Journal of Heuristics, Springer, vol. 27(6), pages 1021-1055, December.
    20. Kallestad, Jakob & Hasibi, Ramin & Hemmati, Ahmad & Sörensen, Kenneth, 2023. "A general deep reinforcement learning hyperheuristic framework for solving combinatorial optimization problems," European Journal of Operational Research, Elsevier, vol. 309(1), pages 446-468.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:spr:compst:v:34:y:2019:i:3:d:10.1007_s00180-018-0832-9. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Sonal Shukla or Springer Nature Abstracting and Indexing (email available below). General contact details of provider: http://www.springer.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.