Regression Model Based on Sparse Bayesian Learning
Yan Li, Fang Liu, Lei Yu, Quan Qi · 2010
Sparse Bayesian learning (SBL) and relevance vector machines(RVM) have received much attention in the machine learning, which as a means of achieving regression. The methodology relies on a parameterized prior that encourages models with few non-zero weights. In this paper, we present a new and efficient algorithm which exploits properties of the marginal likelihood function to enable maximisation via a principled and efficient sequential addition and deletion of candidate basis functions. Meanwhile, regression model has been built based on this algorithm.