Learning with Fredholm kernel has attracted increasing attention recentlysince it can effectively utilize the data information to improve the predictionperformance. Despite rapid progress on theoretical and experimentalevaluations, its generalization analysis has not been explored in learningtheory literature. In this paper, we establish the generalization bound ofleast square regularized regression with Fredholm kernel, which implies thatthe fast learning rate O(l^{-1}) can be reached under mild capacity conditions.Simulated examples show that this Fredholm regression algorithm can achieve thesatisfactory prediction performance.
展开▼
