In some learning approach problems, the learning algorithm receives training and test samples drawn according to the same distribution. However, this assumption is not available in practice. In face super-resolution, when the training sample availabl