Wu, Si, Feng, Jianfeng and Amari, Shun-ichi (2006) The ideal noisy environment for fast neural computation. In: 3rd International Symposium on Neural Networks (ISSN 2006), Chengdu, PEOPLES R CHINA, MAY 28-31, 2006. Published in: ADVANCES IN NEURAL NETWORKS - ISNN 2006, PT 1, 3971 (Part 1). pp. 1-6.

Full text not available from this repository.

Abstract

A central issue in computational neuroscience is to answer why neural systems can process information extremely fast. Here we investigate the effect of noise and the collaborative activity of a neural population on speeding up computation. We find that 1) when input noise is Poissonian, i.e., its variance is proportional to the mean, and 2) when the neural ensemble is initially at its stochastic equilibrium state, noise has the 'best' effect of accelerating computation, in the sense that the input strength is linearly encoded by the number of neurons firing in a short-time window, and that the neural system can use a simple strategy to read out the stimulus rapidly and accurately.

Item Type:	Conference Item (UNSPECIFIED)
Subjects:	Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software
Divisions:	Faculty of Science > Centre for Scientific Computing Faculty of Science > Computer Science
Series Name:	LECTURE NOTES IN COMPUTER SCIENCE
Journal or Publication Title:	ADVANCES IN NEURAL NETWORKS - ISNN 2006, PT 1
Publisher:	SPRINGER-VERLAG BERLIN
ISBN:	3-540-34439-X
ISSN:	0302-9743
Editor:	Wang, J and Yi, Z and Zurada, JM and Lu, BL and Yin, HJ
Date:	2006
Volume:	3971
Number:	Part 1
Number of Pages:	6
Page Range:	pp. 1-6
Publication Status:	Published
Title of Event:	3rd International Symposium on Neural Networks (ISSN 2006)
Location of Event:	Chengdu, PEOPLES R CHINA
Date(s) of Event:	MAY 28-31, 2006
URI:	http://wrap.warwick.ac.uk/id/eprint/33321

Data sourced from Thomson Reuters' Web of Knowledge

Request changes to a record

Actions (login required)

View Item