[Comp-neuro] New paper on optimal sigmoidal tuning curves

[Mark McDonnell]

Dear Colleagues,

A new paper on information theoretically optimal sigmoidal tuning curves has been published in Physical Review Letters:

McDonnell M D and Stocks N G, "Maximally Informative Stimuli and Tuning Curves for Sigmoidal Rate-Coding Neurons and Populations," Physical Review Letters 101, 058103 (2008)

Abstract: 

A general method for deriving maximally informative sigmoidal tuning curves for neural systems with small normalized variability is presented. The optimal tuning curve is a nonlinear function of the cumulative distribution function of the stimulus and depends on the mean-variance relationship of the neural system. The derivation is based on a known relationship between Shannon's mutual information and Fisher information, and the optimality of Jeffrey's prior. It relies on the existence of closed-form solutions to the converse problem of optimizing the stimulus distribution for a given tuning curve. It is shown that maximum mutual information corresponds to constant Fisher information only if the stimulus is uniformly distributed. As an example, the case of sub-Poisson binomial firing statistics is analyzed in detail.

DOI: http://dx.doi.org/10.1103/PhysRevLett.101.058103

Arxiv Preprint: http://arxiv.org/abs/0802.1570

Virtual Journal of Biological Physics Research:  http://scitation.aip.org/getabs/servlet/GetabsServlet?prog=normal&id=VIRT02000016000003000103000001&idtype=cvips&gifs=Yes

Regards

Dr Mark McDonnell
 
Research Fellow

Institute for Telecommunications Research
University of South Australia
SPRI Building
Mawson Lakes Boulevard
Mawson Lakes SA 5095 AUSTRALIA

Phone: +61 8 8302 3341 
Fax: +61 8302 3817

URL:  http://people.unisa.edu.au/Mark.McDonnell
Email: mark.mcdonnell at unisa.edu.au