CrossNets: Neuromorphic networks for nanoelectronic implementation. (English) Zbl 1037.68723

Kaynak, Okyay (ed.) et al., Artificial neural networks and neural information processing — ICANN/ICONIP 2003. Joint international conference ICANN/ICONIP 2003, Istanbul, Turkey, 26–29, 2003. Proceedings. Berlin: Springer (ISBN 3-540-40408-2/pbk). Lect. Notes Comput. Sci. 2714, 753-760 (2003).
Summary: Hybrid “CMOL” integrated circuits, incorporating advanced CMOS devices for neural cell bodies, nanowires as axons and dendrites, and single-molecule latching switches as synapses, may be used for the hardware implementation of extremely dense (\(\sim 10^7\) cells and \(\sim 10^{12}\) synapses per \(\text{cm}^{2})\) neuromorphic networks, operating up to \(10^{6}\) times faster than their biological prototypes. We are exploring several “CrossNet” architectures that accommodate the limitations imposed by CMOL hardware and should allow effective training of the networks without a direct external access to individual synapses. CrossNet training in the Hopfield mode have been confirmed on a software model of the network.
For the entire collection see [Zbl 1029.00055].


68T05 Learning and adaptive systems in artificial intelligence
92B20 Neural networks for/in biological studies, artificial life and related topics


Full Text: Link