Coulomb Classifiers: Reinterpreting SVMs as Electrostatic Systems ; CU-CS-921-01

Coulomb Classifiers: Generalizing Support Vector Machines via an Analogy to Electrostatic Systems

We introduce a family of classifiers based on a physical analogy to an electrostatic system of charged conductors. The family, called Coulomb classifiers, includes the two best-known support-vector machines (SVMs), the ν–SVM and the C–SVM. In the electrostatics analogy, a training example corresponds to a charged conductor at a given location in space, the classification function corresponds to...

Coulomb Classi ers: Generalizing Support Vector Machines via an Analogy to Electrostatic Systems

We introduce a family of classi ers based on a physical analogy to an electrostatic system of charged conductors. The family, called Coulomb classi ers, includes the two best-known support-vector machines (SVMs), the {SVM and the C{SVM. In the electrostatics analogy, a training example corresponds to a charged conductor at a given location in space, the classi cation function corresponds to the...

Linear Time Nonparametric Classification and Feature Selection with Polynomial MPMC Cascades for Large Datasets ; CU-CS-977-04

The recently proposed Polynomial MPMC Cascade (PMC) algorithm is a nonparametric classifier for high-dimensional non-linear binary classification with performance competitive with state-of-the-art classifiers like SVMs. Importantly, the algorithm has linear-time complexity with respect to both training-set size and dimensionality of the problem. In this paper, we show how we can exploit this co...

A Restricted Additive Schwarz Preconditioner with Harmonic Overlap for Symmetric Positive Definite Linear Systems ; CU-CS-920-01

Fingerprint classification using one-vs-all support vector machines dynamically ordered with naive Bayes classifiers

Fingerprint classification reduces the number of possible matches in automated fingerprint identification systems by categorizing fingerprints into predefined classes. Support vector machines (SVMs) are widely used in pattern classification and have produced high accuracy when performing fingerprint classification. In order to effectively apply SVMs to multi-class fingerprint classification sys...