A general framework for interpretable neural learning based on local information-theoretic goal functions
Proceedings of the National Academy of Sciences · 11.03.2025 · 122 (10)
DOI: 10.1073/pnas.2408125122 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 05.03.2025. Verfasst von Abdullah Makkeh und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Data-driven Analysis of Biological Networks, Göttingen Campus Institute for Dynamics of Biological Networks, University of Göttingen, G…. Hauptthemen laut MEDLINE: Neural Networks, Computer, Learning, Models, Neurological, Information Theory. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 85 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Volkswagen Foundation (VolkswagenStiftung), European Commission (EC), Deutsche Forschungsgemeinschaft (DFG).
Abstract (Originalsprache)
Despite the impressive performance of biological and artificial networks, an intuitive understanding of how their local learning dynamics contribute to network-level task solutions remains a challenge to this date. Efforts to bring learning to a more local scale indeed lead to valuable insights, however, a general constructive approach to describe local learning goals that is both interpretable and adaptable across diverse tasks is still missing. We have previously formulated a local information processing goal that is highly adaptable and interpretable for a model neuron with compartmental structure. Building on recent advances in Partial Information Decomposition (PID), we here derive a corresponding parametric local learning rule, which allows us to introduce “infomorphic” neural networks. We demonstrate the versatility of these networks to perform tasks from supervised, unsupervised, and memory learning. By leveraging the interpretable nature of the PID framework, infomorphic networks represent a valuable tool to advance our understanding of the intricate structure of local learning.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Data-driven Analysis of Biological Networks, Göttingen Campus Institute for Dynamics of Biological Networks, University of Göttingen, Göttingen 37077, Germany.
- Complex Systems Theory, Max Planck Institute for Dynamics and Self-Organization, Göttingen 37077, Germany.
- Department of Chemistry and Applied Biosciences, ETH Zurich, Zurich 8092, Switzerland.
- University of Göttingen, Göttingen 37073, Germany.
- Cluster of Excellence "Multiscale Bioimaging: from Molecular Machines to Networks of Excitable Cells" (MBExC), University of Göttingen, Göttingen 37073, Germany.
ORCID-Profile: Abdullah Makkeh, Marcel Graetz, Andreas C Schneider, David A Ehrlich, Viola Priesemann, Michael Wibral
Förderung
- Volkswagen Foundation (VolkswagenStiftung)
- European Commission (EC)
- Deutsche Forschungsgemeinschaft (DFG)
- Niedersächsisches Ministerium für Wissenschaft und Kultur (MWK)
Zitieren
APA
Makkeh, A., Graetz, M., Schneider, A. C., Ehrlich, D. A., Priesemann, V., & Wibral, M. (2025). A general framework for interpretable neural learning based on local information-theoretic goal functions. Proceedings of the National Academy of Sciences, 122(10). https://doi.org/10.1073/pnas.2408125122
Vancouver
Makkeh A, Graetz M, Schneider AC, Ehrlich DA, Priesemann V, Wibral M. A general framework for interpretable neural learning based on local information-theoretic goal functions. Proceedings of the National Academy of Sciences. 2025;122(10). doi:10.1073/pnas.2408125122
BibTeX
@article{Makkeh2025,
author = {Makkeh, Abdullah and Graetz, Marcel and Schneider, Andreas C and Ehrlich, David A and Priesemann, Viola and Wibral, Michael},
title = {A general framework for interpretable neural learning based on local information-theoretic goal functions},
journal = {Proceedings of the National Academy of Sciences},
year = {2025},
volume = {122},
number = {10},
doi = {10.1073/pnas.2408125122}
}
EUVIMED