Relevance vector machine
In mathematics, a Relevance Vector Machine (RVM) is a machine learning technique that uses Bayesian inference to obtain parsimonious solutions for regression and probabilistic classification. A greedy optimisation procedure and thus fast version were subsequently developed.