Regularized Least Square Regression with Unbounded and Dependent Sampling

Abstract

This paper mainly focuses on the least square regression problem for the\n$\\alpha $ -mixing and $\\varphi $ -mixing processes. The standard bound assumption for output\ndata is abandoned and the learning algorithm is implemented with\nsamples drawn from dependent sampling process with a more general\noutput data condition. Capacity independent error bounds and learning\nrates are deduced by means of the integral operator technique.