Penalized regression splines in Mixture Density Networks

Seifert Quentin Edward (), Thielmann Anton, Bergherr Elisabeth, Säfken Benjamin, Zierk Jakob, Rauh Manfred and Hepp Tobias
Additional contact information
Seifert Quentin Edward: Chair of Spatial Data Science and Statistical Learning, University of Göttingen, Göttingen, Germany
Thielmann Anton: Chair of Data Science and Applied Statistics, TU Clausthal, Clausthal, Germany
Bergherr Elisabeth: Chair of Spatial Data Science and Statistical Learning, University of Göttingen, Göttingen, Germany
Säfken Benjamin: Chair of Data Science and Applied Statistics, TU Clausthal, Clausthal, Germany
Zierk Jakob: Department of Paediatrics and Adolescent Medicine, University Hospital Erlangen, Erlangen, Germany
Rauh Manfred: Department of Paediatrics and Adolescent Medicine, University Hospital Erlangen, Erlangen, Germany
Hepp Tobias: Chair of Spatial Data Science and Statistical Learning, University of Göttingen, Göttingen, Germany

The International Journal of Biostatistics, 2025, vol. 21, issue 1, 239-253

Abstract: Mixture Density Networks (MDN) belong to a class of models that can be applied to data which cannot be sufficiently described by a single distribution since it originates from different components of the main unit and therefore needs to be described by a mixture of densities. In some situations, MDNs may have problems with the proper identification of the latent components. While these identification issues can to some extent be contained by using custom initialization strategies for the network weights, this solution is still less than ideal since it involves subjective opinions. We therefore suggest replacing the hidden layers between the model input and the output parameter vector of MDNs and estimating the respective distributional parameters with penalized cubic regression splines. Results on simulated data from both Gaussian and Gamma mixture distributions motivated by an application to indirect reference interval estimation drastically improved the identification performance with all splines reliably converging to their true parameter values.

Keywords: neural networks; distributional regression; finite mixture models; regression splines (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://doi.org/10.1515/ijb-2023-0134 (text/html)
For access to full text, subscription to the journal or payment for the individual article is required.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:bpj:ijbist:v:21:y:2025:i:1:p:239-253:n:1004

Ordering information: This journal article can be ordered from
https://www.degruyte ... journal/key/ijb/html

DOI: 10.1515/ijb-2023-0134

Access Statistics for this article

The International Journal of Biostatistics is currently edited by Antoine Chambaz, Alan E. Hubbard and Mark J. van der Laan

More articles in The International Journal of Biostatistics from De Gruyter
Bibliographic data for series maintained by Peter Golla ().