EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Targeting operational regimes of interest in recurrent neural networks

Pierre Ekelmans iD, Nataliya Kraynyukova, Tatjana Tchumatchenko iD

PLOS Computational Biology · 2023

DOI: 10.1371/journal.pcbi.1011097 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, 2023. Verfasst von Pierre Ekelmans und 2 weiteren Autor:innen. Beteiligte Einrichtung: Theory of Neural Dynamics group, Max Planck Institute for Brain Research, Frankfurt am Main, Germany. Hauptthemen laut MEDLINE: Nerve Net, Neural Networks, Computer. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 87 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Max-Planck-Gesellschaft, Medizinische Fakultät, Rheinische Friedrich-Wilhelms-Universität Bonn, Universitätsmedizin der Johannes Gutenberg-Universität Mainz.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Neural computations emerge from local recurrent neural circuits or computational units such as cortical columns that comprise hundreds to a few thousand neurons. Continuous progress in connectomics, electrophysiology, and calcium imaging require tractable spiking network models that can consistently incorporate new information about the network structure and reproduce the recorded neural activity features. However, for spiking networks, it is challenging to predict which connectivity configurations and neural properties can generate fundamental operational states and specific experimentally reported nonlinear cortical computations. Theoretical descriptions for the computational state of cortical spiking circuits are diverse, including the balanced state where excitatory and inhibitory inputs balance almost perfectly or the inhibition stabilized state (ISN) where the excitatory part of the circuit is unstable. It remains an open question whether these states can co-exist with experimentally reported nonlinear computations and whether they can be recovered in biologically realistic implementations of spiking networks. Here, we show how to identify spiking network connectivity patterns underlying diverse nonlinear computations such as XOR, bistability, inhibitory stabilization, supersaturation, and persistent activity. We establish a mapping between the stabilized supralinear network (SSN) and spiking activity which allows us to pinpoint the location in parameter space where these activity regimes occur. Notably, we find that biologically-sized spiking networks can have irregular asynchronous activity that does not require strong excitation-inhibition balance or large feedforward input and we show that the dynamic firing rate trajectories in spiking networks can be precisely targeted without error-driven training algorithms.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Nerve NetNeural Networks, ComputerAction PotentialsNeuronsAlgorithmsModels, NeurologicalNeural Inhibition

Autor:innen und Institutionen

  • Theory of Neural Dynamics group, Max Planck Institute for Brain Research, Frankfurt am Main, Germany.
  • Frankfurt Institute for Advanced Studies, Frankfurt am Main, Germany.
  • Institute of Experimental Epileptology and Cognition Research, Life and Brain Center, Universitätsklinikum Bonn, Bonn, Germany.
  • Institute of physiological chemistry, Medical center of the Johannes Gutenberg-University Mainz, Mainz, Germany.

ORCID-Profile: Pierre Ekelmans, Tatjana Tchumatchenko

Förderung

  • Max-Planck-Gesellschaft
  • Medizinische Fakultät, Rheinische Friedrich-Wilhelms-Universität Bonn
  • Universitätsmedizin der Johannes Gutenberg-Universität Mainz
  • Deutsche Forschungsgemeinschaft
  • Hessisches Ministerium für Wissenschaft und Kunst

Zitieren

APA

Ekelmans, P., Kraynyukova, N., & Tchumatchenko, T. (2023). Targeting operational regimes of interest in recurrent neural networks. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1011097

Vancouver

Ekelmans P, Kraynyukova N, Tchumatchenko T. Targeting operational regimes of interest in recurrent neural networks. PLOS Computational Biology. 2023. doi:10.1371/journal.pcbi.1011097

BibTeX
@article{Ekelmans2023,
  author = {Ekelmans, Pierre and Kraynyukova, Nataliya and Tchumatchenko, Tatjana},
  title = {Targeting operational regimes of interest in recurrent neural networks},
  journal = {PLOS Computational Biology},
  year = {2023},
  doi = {10.1371/journal.pcbi.1011097}
}
Kostenlos starten →