Open AccessTheory of a superconducting artificial neuron for extended backpropagation learning algorism
Author(s) -
Haruna Katayama,
Toshiyuki Fujii,
Noriyuki Hatakenaka
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i3.29.18551
Subject(s) - algorism , squid , superconductivity , backpropagation , artificial neuron , artificial neural network , multilayer perceptron , sigmoid function , josephson effect , physics , condensed matter physics , materials science , computer science , artificial intelligence , mathematics , biology , ecology , arithmetic
An artificial neuron using superconducting devices, so-called double SQUID, applicable to the extended backpropagation learning algorism is studied. It is shown that the tunable slope of the sigmoid function required in the algorism can be achieved under the fixed temperature by externally applied magnetic fields threading the ring with two Josephson junctions in the double SQUID.