2013/12/09 by Matteo Colangeli, Francesco Rugiano, Eros Pasero
Computer Science · Mathematics · Physics and Astronomy · #Algorithm #Artificial intelligence #Computer science #Data mining #Face and Expression Recognition #Image (mathematics) #Machine learning #Mathematics #Neural Networks and Applications #Observable #Pattern recognition (psychology) #Perspective (graphical) #Physics #Resolution (logic) #Scale (ratio) #Set (abstract data type) #Statistical Mechanics and Entropy #Statistical analysis #Statistical learning #Statistical mechanics #Statistical physics #Statistical theory #Statistics #Support vector machine #cond-mat.dis-nn #cond-mat.stat-mech
paper · pdf · doi:10.1016/j.chaos.2013.10.006
openalex publication_date 2013/12/09 · arxiv created 2014/04/09 · arxiv updated 2015/06/19 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
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.