Mercer Large-Scale Kernel Machines from Ridge Function Perspective

To present Mercer large-scale kernel machines from a ridge function perspective, we recall the results by Lin and Pinkus from Fundamentality of ridge functions. We consider the main theorem of the recent paper by Rachimi and Recht, 2008, Random features for large-scale kernel machines in terms of the Approximation Theory. We study which kernels can be approximated by a sum of cosine function products with arguments depending on $x$ and $y$ and present the obstacles of such an approach. The results of this article may have various applications in Deep Learning, especially in problems related to Image Processing.