EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks

Basile Confavreux iD, Everton J Agnes iD, Friedemann Zenke iD, Henning Sprekeler iD, Tim P Vogels iD

PLOS Computational Biology · 2025

DOI: 10.1371/journal.pcbi.1012910 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, 2025. Verfasst von Basile Confavreux und 4 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Science and Technology Austria, Klosterneuburg, Austria. Hauptthemen laut MEDLINE: Neuronal Plasticity, Models, Neurological, Action Potentials, Nerve Net. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Wellcome Trust, HORIZON EUROPE European Research Council.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Synaptic plasticity is a key player in the brain’s life-long learning abilities. However, due to experimental limitations, the mechanistic link between synaptic plasticity rules and the network-level computations they enable remain opaque. Here we use evolutionary strategies (ES) to meta learn local co-active plasticity rules in large recurrent spiking networks with excitatory (E) and inhibitory (I) neurons, using parameterizations of increasing complexity. We discover rules that robustly stabilize network dynamics for all four synapse types acting in isolation (E-to-E, E-to-I, I-to-E and I-to-I). More complex functions such as familiarity detection can also be included in the search constraints. However, our meta learning strategy begins to fail for co-active rules of increasing complexity, as it is challenging to devise loss functions that effectively constrain network dynamics to plausible solutions a priori . Moreover, in line with previous work, we can find multiple degenerate solutions with identical network behaviour. As a local optimization strategy, ES provides one solution at a time and makes exploration of this degeneracy cumbersome. Regardless, we can glean the interdependecies of various plasticity parameters by considering the covariance matrix learned alongside the optimal rule with ES. Our work provides a proof of principle for the success of machine-learning-guided discovery of plasticity rules in large spiking networks, and points at the necessity of more elaborate search strategies going forward.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neuronal PlasticityModels, NeurologicalAction PotentialsNerve NetLearningComputational BiologyNeuronsHumansAnimalsComputer SimulationSynapses

Autor:innen und Institutionen

  • Institute of Science and Technology Austria, Klosterneuburg, Austria.
  • Gatsby Computational Neuroscience Unit, University College London, London, United Kingdom.
  • Biozentrum, University of Basel, Basel, Switzerland.
  • Friedrich Miescher Institute for Biomedical Research, Basel, Switzerland.
  • Technische Universität Berlin, Berlin, Germany.

ORCID-Profile: Basile Confavreux, Everton J Agnes, Friedemann Zenke, Henning Sprekeler, Tim P Vogels

Förderung

  • Wellcome Trust
  • HORIZON EUROPE European Research Council

Zitieren

APA

Confavreux, B., Agnes, E. J., Zenke, F., Sprekeler, H., & Vogels, T. P. (2025). Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1012910

Vancouver

Confavreux B, Agnes EJ, Zenke F, Sprekeler H, Vogels TP. Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks. PLOS Computational Biology. 2025. doi:10.1371/journal.pcbi.1012910

BibTeX
@article{Confavreux2025,
  author = {Confavreux, Basile and Agnes, Everton J and Zenke, Friedemann and Sprekeler, Henning and Vogels, Tim P},
  title = {Balancing complexity, performance and plausibility to meta learn plasticity rules in recurrent spiking networks},
  journal = {PLOS Computational Biology},
  year = {2025},
  doi = {10.1371/journal.pcbi.1012910}
}
Kostenlos starten →