Information loss in an optimal maximum likelihood decoding
Biological Physics
2007-05-23 v1 Disordered Systems and Neural Networks
Data Analysis, Statistics and Probability
Quantitative Methods
Abstract
The mutual information between a set of stimuli and the elicited neural responses is compared to the corresponding decoded information. The decoding procedure is presented as an artificial distortion of the joint probabilities between stimuli and responses. The information loss is quantified. Whenever the probabilities are only slightly distorted, the information loss is shown to be quadratic in the distortion
Keywords
Cite
@article{arxiv.physics/0110074,
title = {Information loss in an optimal maximum likelihood decoding},
author = {Ines Samengo},
journal= {arXiv preprint arXiv:physics/0110074},
year = {2007}
}
Comments
5 pages, no figures, to be published in Neural Computation