Pattern recognition at different scales: a statistical perspective
Statistical Mechanics
2015-06-19 v1 Disordered Systems and Neural Networks
Abstract
In this paper we borrow concepts from Information Theory and Statistical Mechanics to perform a pattern recognition procedure on a set of x-ray hazelnut images. We identify two relevant statistical scales, whose ratio affects the performance of a machine learning algorithm based on statistical observables, and discuss the dependence of such scales on the image resolution. Finally, by averaging the performance of a Support Vector Machines algorithm over a set of training samples, we numerically verify the predicted onset of an optimal scale of resolution, at which the pattern recognition is favoured.
Cite
@article{arxiv.1404.2638,
title = {Pattern recognition at different scales: a statistical perspective},
author = {Matteo Colangeli and Francesco Rugiano and Eros Pasero},
journal= {arXiv preprint arXiv:1404.2638},
year = {2015}
}