This paper provides a new look at radial basis function regression that reveals striking similarities with the traditional optimal experimental design framework. We show theoreti- cally 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 net- work, 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 sta- tistical 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.
Fokoue, Ernest and Goel, Prem, "An optimal experimental design perspective on redial basis function regression" (2010). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus