Hardware-oriented neuron modeling approach by reconfigurable asynchronous cellar automaton. (English) Zbl 1315.92006
Ohira, Toru (ed.) et al., Mathematical approaches to biological systems. Networks, oscillations, and collective motions. Tokyo: Springer (ISBN 978-4-431-55443-1/hbk; 978-4-431-55444-8/ebook). 55-75 (2015).
Summary: A variety of neuron models have been presented so far, where there exist two major modeling approaches: a nonlinear ordinary differential equation approach and a nonlinear difference equation approach. On the other hand, recently, our group has been developing a new hardware-oriented neuron modeling approach: a reconfigurable asynchronous cellar automaton approach. In this chapter, neuron-like bifurcations of the asynchronous cellar automaton neuron model are analyzed and an on-chip learning algorithm (on-chip dynamic circuit reconfiguration algorithm) for reproducing biological neuron’s behaviors is demonstrated.
For the entire collection see [Zbl 1358.92006].
For the entire collection see [Zbl 1358.92006].
MSC:
92B20 | Neural networks for/in biological studies, artificial life and related topics |
68T05 | Learning and adaptive systems in artificial intelligence |