Kernel Autoassociator with Applications to Visual Classification
Haihong Zhang, Weimin Huang, Zhiyong Huang, Bailing Zhang · 2015
Autoassociator is an important issue in concept learn-ing, and the learned concept of a particular class can be used to distinguish the class from the others. For nonlinear autoassociation, this paper presents a new model referred to as kernel autoassociator. Using kernel feature space as a potential nonlinear manifold, the model formulates the au-toassociation as a special reconstruction problem from ker-nel feature space to input space. Two methods are devel-oped to solve the problem. We evaluate the autoassociator with artificial data, and apply it to handwritten digit recog-nition and multiview face recognition, yielding positive ex-perimental results. 1.