• Home
  1. Home
  2. Insights from a simple expression for linear fisher information in a recurrently connected population of spiking neurons
Scientific article
OA Policy
English

Insights from a simple expression for linear fisher information in a recurrently connected population of spiking neurons

Contributors

Beck, Jeffrey; Bejjanki, Vikranth R; Pouget, Alexandre

Published in

Neural computation, vol. 23, no. 6, p. 1484-1502

Publication date

2011

Abstract

A simple expression for a lower bound of Fisher information is derived for a network of recurrently connected spiking neurons that have been driven to a noise-perturbed steady state. We call this lower bound linear Fisher information, as it corresponds to the Fisher information that can be recovered by a locally optimal linear estimator. Unlike recent similar calculations, the approach used here includes the effects of nonlinear gain functions and correlated input noise and yields a surprisingly simple and intuitive expression that offers substantial insight into the sources of information degradation across successive layers of a neural network. Here, this expression is used to (1) compute the optimal (i.e., information-maximizing) firing rate of a neuron, (2) demonstrate why sharpening tuning curves by either thresholding or the action of recurrent connectivity is generally a bad idea, (3) show how a single cortical expansion is sufficient to instantiate a redundant population code that can propagate across multiple cortical layers with minimal information loss, and (4) show that optimal recurrent connectivity strongly depends on the covariance structure of the inputs to the network.

Keywords

  • Action Potentials/physiology
  • Models, Neurological
  • Nerve Net/physiology
  • Neural Networks (Computer)
  • Neurons/physiology

UNIGE affiliation entities

UNIGE research groups

Citation (ISO format)

BECK, Jeffrey, BEJJANKI, Vikranth R, POUGET, Alexandre. Insights from a simple expression for linear fisher information in a recurrently connected population of spiking neurons. In: Neural computation, 2011, vol. 23, n° 6, p. 1484–1502. doi: 10.1162/NECO_a_00125

Main files (1)

Article (Published version)

Identifiers

Journal ISSN

0899-7667
713views
902downloads

Technical informations

Creation

19/11/2012 14:53:00

First validation

19/11/2012 14:53:00

Update

14/03/2023 17:49:04

Status update

14/03/2023 17:49:04

Last indexation

30/10/2024 08:11:30
All rights reserved by Archive ouverte UNIGE and the University of Geneva
OpenAIRE Indexed & Validated