EUVIMEDEUVIMED entdecken
Crossref

Similarity Matching Networks: Hebbian Learning and Convergence Over Multiple Timescales

Veronica Centorrino, Francesco Bullo, Giovanni Russo

Neural Computation · 30.04.2026 · 38 (5) · 725-764

DOI: 10.1162/neco.a.1509 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Neural Computation, online erschienen am 30.04.2026. Verfasst von Veronica Centorrino und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information and Electric Engineering and Applied Mathematics, University of Salerno, 84084 Fisciano SA, Italy vcentorrino@control.ee.et…. Die Arbeit nennt 45 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A recent breakthrough in biologically plausible normative frameworks for dimensionality reduction is based on the similarity matching cost function and the low-rank matrix approximation problem. Despite clear biological interpretation, successful application in several domains, and experimental validation, a formal complete convergence analysis remains elusive. Building on this framework, we consider and analyze a continuous-time neural network, the similarity matching network, for principal subspace projection. Derived from a min-max-min objective, this biologically plausible network consists of three coupled dynamics evolving at different timescales: neural dynamics, lateral synaptic dynamics, and feedforward synaptic dynamics at the fast, intermediate, and slow timescales, respectively. The feedforward and lateral synaptic dynamics consist of Hebbian and anti-Hebbian learning rules, respectively. By leveraging a multilevel optimization framework, we prove convergence of the dynamics in the offline setting. Specifically, at the first level (fast timescale), we show strong convexity of the cost function and global exponential convergence of the corresponding gradient-flow dynamics. At the second level (intermediate timescale), we prove strong concavity of the cost function and exponential convergence of the corresponding gradient-flow dynamics within the space of positive definite matrices. At the third and final level (slow timescale), we study a nonconvex and nonsmooth cost function, provide explicit expressions for its global minima, and prove almost sure convergence of the corresponding gradient-flow dynamics to the global minima. These results rely on two empirically motivated conjectures that are supported by thorough numerical experiments. Finally, we validate the effectiveness of our approach via a numerical example.

Autor:innen und Institutionen

  • Department of Information and Electric Engineering and Applied Mathematics, University of Salerno, 84084 Fisciano SA, Italy vcentorrino@control.ee.ethz.ch.
  • Center for Control, Dynamical Systems, and Computation, UC Santa Barbara, CA 93106, USA bullo@ucsb.edu.
  • Department of Information and Electric Engineering and Applied Mathematics, University of Salerno, 84084 Fisciano SA, Italy giovarusso@unisa.it.

Zitieren

APA

Centorrino, V., Bullo, F., & Russo, G. (2026). Similarity Matching Networks: Hebbian Learning and Convergence Over Multiple Timescales. Neural Computation, 38(5), 725-764. https://doi.org/10.1162/neco.a.1509

Vancouver

Centorrino V, Bullo F, Russo G. Similarity Matching Networks: Hebbian Learning and Convergence Over Multiple Timescales. Neural Computation. 2026;38(5):725-764. doi:10.1162/neco.a.1509

BibTeX
@article{Centorrino2026,
  author = {Centorrino, Veronica and Bullo, Francesco and Russo, Giovanni},
  title = {Similarity Matching Networks: Hebbian Learning and Convergence Over Multiple Timescales},
  journal = {Neural Computation},
  year = {2026},
  volume = {38},
  number = {5},
  pages = {725--764},
  doi = {10.1162/neco.a.1509}
}
Kostenlos starten →