Machine learning for non-metric proximity data

This blog provides literature, algorithms and data sets for the analysis of (indefinite) proximity data. In machine learning kernels are given as proximity data. But if the proximity measure is non-metric most kernel approaches are inaccurate or fail. This blog shows ways how to deal with these so called indefinite, non-positive or non-psd proximity data, providing links to literature and algorithms. The final objective is to provide - Probabilistic Models in Pseudo-Euclidean Spaces (ProMoS)

Friday, 5 January 2018

Code for indefinite Core Vector Machine (iCVM) published

A simplified matlab code and an armadillo/C++ implementation of the indefinite
core vector machine (iCVM) is published at iCVM - Indefinite Core Vector Machine
and MLOSS it implements ideas published in the paper

Frank-Michael Schleif, Peter Tiño:
Indefinite Core Vector Machine. Pattern Recognition 71: 187-195 (2017)

The C++ code provides additionally a new approach to (re-)sparsify the indefinite model
such that the final indefinite decision function remains sparse and permits easy out of sample extension.