CMOL CrossNets as Pattern Classifiers

This presentation has two goals: (i) to review the recently suggested concept of bio-inspired CrossNet architectures for future hybrid CMOL VLSI circuits and (ii) to present new results concerning the prospects and problems of using these neuromorphic networks as classifiers of very large patterns, in particular of high-resolution optical images. We show that the unparalleled density and speed of CMOL circuits may enable to perform such important and challenging tasks as, for example, online recognition of a face in a highresolution image of a large crowd. 1 CrossNets There is a growing consensus that the forthcoming problems of the Moore Law [1] may be only resolved by the transfer from a purely semiconductor-transistor (CMOS) technology to hybrid (“CMOL”) integrated circuits [2, 3]. Such circuit would complement a CMOS chip with a nanowire crossbar (Fig. 1) with nanodevices (e.g., specially designed functional molecules) formed between the nanowires at each crosspoint.

• Type: Conference
• Year: 2005
• Where: IWANN
• Authors: Jung Hoon Lee, Konstantin Likharev
• Keywords: Bio-inspired Crossnet Architectures, CMOL Circuits, CMOL VLSI Circuits, IWANN 2005