Then a convex set is assigned to the parameter vector, and the size of the convex set is minimized such that every possible data point can be predicted by one possible value of the parameters.Įllipsoidal parameters sets were used by Campi (2009), which yield a convex optimization program to train the IPM.
Usually the basis is made up of polynomial features or a radial basis is sometimes used. Typically the interval predictor model is created by specifying a parametric function, which is usually chosen to be the product of a parameter vector and a basis.