A Proposal of TAM Network with Extensive Receptive Field

Abstract

TAM (Topographic Attentive Mapping) network is a biologically-motivated neural network with Gabor function type receptive fields. However, the sturucture of receptive fields is a mono-layer, and there is a lack of performance for rotation characteristic of images. In this paper, we formulate a new TAM network structure with extensive and multilayer receptive fields. We also show the usefulness of TAM nework using some examples of character recognitions.