An Optimal Experimental Design Perspective on Radial Basis Function Regression

作者：Fokoue Ernest^*; Goel Prem

来源：Communications in Statistics - Theory and Methods, 2011, 40(7): 1184-1195.

DOI：10.1080/03610920903564735

• D-Optimality
• Marginal likelihood
• Maximum a posterior (MAP)
• Radial basis function regression
• Relevance vector machine
• Sensor selection
• Sparsity

This article provides a new look at radial basis function regression that reveals striking similarities with the traditional optimal experimental design framework. We show theoretically and computationally that the so-called relevant vectors derived through the relevance vector machine (RVM) and corresponding to the centers of the radial basis function network, are very similar and often identical to the support points obtained through various optimal experimental design criteria like D-optimality. This allows us to provide a statistical meaning to the relevant centers in the context of radial basis function regression, but also opens the door to a variety of ways of approach optimal experimental design in multivariate settings.

• 出版日期2011