Algorithmic adjustment of neural field parameters - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Reports (Research Report) Year : 2009

Algorithmic adjustment of neural field parameters

Frédéric Alexandre
Jérémy Fix
Nicolas P. Rougier
Thierry Viéville

Abstract

Revisiting neural-field calculation maps in the discrete case, we propose algorithmic mechanisms allowing to choose a right set of parameters in order to both (i) guaranty the stability of the calculation and (ii) tune the shape of the output map. These results do not ``prove'' the existence of stable bump solutions, this being already known and extensively verified numerically, but allow to calculate algorithmically the related parameters. The results apply to scalar and vectorial neural-fields thus allowing to bypass the inherent limitations brought by mean frequency models and also take the laminar structure of the cortex or high-level representation of cortical computations into account. We obtain an easy to implement procedure that guaranty the convergence of the map onto a fixed point, even with large sampling steps. Furthermore, we report how rectification is the minimal required non-linearity to obtain usual neural-field behaviors. We also propose a way to control and tune these behaviors (filtering, selection, tracking, remanence) and optimize the convergence rate. This applies to both non parametric profiles, i.e. adjusting the weight values directly, or to parametric profiles and thus adjusting their parameters (e.g. Mexican-hat profiles). Beyond these algorithmic results, the idea of studying neural computations as discrete dynamical systems and not only the discretization of a continuous system is emphasized here. The outcome is shared as an open-source plug-in module, called EnaS (http://enas.gforge.inria.fr), to be used in existing simulation software.
Fichier principal
Vignette du fichier
rapport.pdf (575.78 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

inria-00382533 , version 1 (08-05-2009)

Identifiers

  • HAL Id : inria-00382533 , version 1

Cite

Frédéric Alexandre, Jérémy Fix, Nicolas P. Rougier, Thierry Viéville. Algorithmic adjustment of neural field parameters. [Research Report] RR-6923, INRIA. 2009, pp.26. ⟨inria-00382533⟩
301 View
152 Download

Share

Gmail Facebook X LinkedIn More