Penalized regression splines in Mixture Density Networks

Seifert QE, Thielmann A, Bergherr E, Säfken B, Zierk J, Rauh M, Hepp T (2025)

Publication Type: Journal article

Publication year: 2025

Journal

International Journal of Biostatistics De Gruyter

DOI: 10.1515/ijb-2023-0134

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.

Authors with CRIS profile

Jakob Zierk Interdisziplinäres Zentrum für Klinische Forschung IZKF-Förderlinien Manfred Rauh Department of Paediatrics and Adolescent Medicine

Involved external institutions

Georg-August-Universität Göttingen DE Germany (DE) Technische Universität Clausthal / Clausthal University of Technology DE Germany (DE)

How to cite

APA:

Seifert, Q.E., Thielmann, A., Bergherr, E., Säfken, B., Zierk, J., Rauh, M., & Hepp, T. (2025). Penalized regression splines in Mixture Density Networks. International Journal of Biostatistics. https://doi.org/10.1515/ijb-2023-0134

MLA:

Seifert, Quentin Edward, et al. "Penalized regression splines in Mixture Density Networks." International Journal of Biostatistics (2025).

BibTeX: Download