Title :
Efficient training algorithms for neural networks based on memristive crossbar circuits
Author :
Irina Kataeva;Farnood Merrikh-Bayat;Elham Zamanidoost;Dmitri Strukov
Author_Institution :
Advanced Research Division, DENSO CORPORATION, Komenoki-cho, Nisshin, Japan, 470-0111
fDate :
7/1/2015 12:00:00 AM
Abstract :
We have adapted backpropagation algorithm for training multilayer perceptron classifier implemented with memristive crossbar circuits. The proposed training approach takes into account switching dynamics of a particular, though very typical, type of memristive devices and weight update restrictions imposed by crossbar topology. The simulation results show that for crossbar-based multilayer perceptron with one hidden layer of 300 neurons misclassification rate on MNIST benchmark could be as low as 1.47% and 4.06% for batch and stochastic algorithms, respectively, which is comparable to the best reported results for similar neural networks.
Keywords :
"CMOS integrated circuits","Switches","Magnetic multilayers","Magnetic resonance imaging","Metals","Nonhomogeneous media","Programming"
Conference_Titel :
Neural Networks (IJCNN), 2015 International Joint Conference on
Electronic_ISBN :
2161-4407
DOI :
10.1109/IJCNN.2015.7280785