Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception

Abstract The robust estimation of dynamical hidden features, such as the position of prey, based on sensory inputs is one of the hallmarks of perception. This dynamical estimation can be rigorously formulated by nonlinear Bayesian filtering theory. Recent experimental and behavioral studies have sho...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Anna Kutschireiter, Simone Carlo Surace, Henning Sprekeler, Jean-Pascal Pfister
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2017
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/6ca9bd6f654040eeb7e72038e632eeb2
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:6ca9bd6f654040eeb7e72038e632eeb2
record_format	dspace
spelling	oai:doaj.org-article:6ca9bd6f654040eeb7e72038e632eeb22021-12-02T12:31:59ZNonlinear Bayesian filtering and learning: a neuronal dynamics for perception10.1038/s41598-017-06519-y2045-2322https://doaj.org/article/6ca9bd6f654040eeb7e72038e632eeb22017-08-01T00:00:00Zhttps://doi.org/10.1038/s41598-017-06519-yhttps://doaj.org/toc/2045-2322Abstract The robust estimation of dynamical hidden features, such as the position of prey, based on sensory inputs is one of the hallmarks of perception. This dynamical estimation can be rigorously formulated by nonlinear Bayesian filtering theory. Recent experimental and behavioral studies have shown that animals’ performance in many tasks is consistent with such a Bayesian statistical interpretation. However, it is presently unclear how a nonlinear Bayesian filter can be efficiently implemented in a network of neurons that satisfies some minimum constraints of biological plausibility. Here, we propose the Neural Particle Filter (NPF), a sampling-based nonlinear Bayesian filter, which does not rely on importance weights. We show that this filter can be interpreted as the neuronal dynamics of a recurrently connected rate-based neural network receiving feed-forward input from sensory neurons. Further, it captures properties of temporal and multi-sensory integration that are crucial for perception, and it allows for online parameter learning with a maximum likelihood approach. The NPF holds the promise to avoid the ‘curse of dimensionality’, and we demonstrate numerically its capability to outperform weighted particle filters in higher dimensions and when the number of particles is limited.Anna KutschireiterSimone Carlo SuraceHenning SprekelerJean-Pascal PfisterNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 7, Iss 1, Pp 1-13 (2017)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Anna Kutschireiter Simone Carlo Surace Henning Sprekeler Jean-Pascal Pfister Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
description	Abstract The robust estimation of dynamical hidden features, such as the position of prey, based on sensory inputs is one of the hallmarks of perception. This dynamical estimation can be rigorously formulated by nonlinear Bayesian filtering theory. Recent experimental and behavioral studies have shown that animals’ performance in many tasks is consistent with such a Bayesian statistical interpretation. However, it is presently unclear how a nonlinear Bayesian filter can be efficiently implemented in a network of neurons that satisfies some minimum constraints of biological plausibility. Here, we propose the Neural Particle Filter (NPF), a sampling-based nonlinear Bayesian filter, which does not rely on importance weights. We show that this filter can be interpreted as the neuronal dynamics of a recurrently connected rate-based neural network receiving feed-forward input from sensory neurons. Further, it captures properties of temporal and multi-sensory integration that are crucial for perception, and it allows for online parameter learning with a maximum likelihood approach. The NPF holds the promise to avoid the ‘curse of dimensionality’, and we demonstrate numerically its capability to outperform weighted particle filters in higher dimensions and when the number of particles is limited.
format	article
author	Anna Kutschireiter Simone Carlo Surace Henning Sprekeler Jean-Pascal Pfister
author_facet	Anna Kutschireiter Simone Carlo Surace Henning Sprekeler Jean-Pascal Pfister
author_sort	Anna Kutschireiter
title	Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
title_short	Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
title_full	Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
title_fullStr	Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
title_full_unstemmed	Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception
title_sort	nonlinear bayesian filtering and learning: a neuronal dynamics for perception
publisher	Nature Portfolio
publishDate	2017
url	https://doaj.org/article/6ca9bd6f654040eeb7e72038e632eeb2
work_keys_str_mv	AT annakutschireiter nonlinearbayesianfilteringandlearninganeuronaldynamicsforperception AT simonecarlosurace nonlinearbayesianfilteringandlearninganeuronaldynamicsforperception AT henningsprekeler nonlinearbayesianfilteringandlearninganeuronaldynamicsforperception AT jeanpascalpfister nonlinearbayesianfilteringandlearninganeuronaldynamicsforperception
_version_	1718394219298553856

Nonlinear Bayesian filtering and learning: a neuronal dynamics for perception

Ejemplares similares