Implementation of multilayer perceptron network with highly uniform passive memristive crossbar circuits

TLDR

Advances in neural computation depend on more efficient hardware, and passive memristive crossbar circuits promise dramatic performance gains, but immature memristor technology limits their functionality. The study demonstrates a one‑hidden‑layer perceptron classifier implemented entirely in mixed‑signal hardware using two passive 20×20 metal‑oxide memristive crossbar arrays integrated with conventional components. The classifier uses two passive 20×20 metal‑oxide memristive crossbar arrays integrated on board with discrete conventional components. The network achieves classification fidelity within 3 % of simulation results while its hardware complexity is nearly ten times higher than prior passive memristive classifiers, thanks to fabrication improvements that reduce memristor I‑V variation.

Abstract

The progress in the field of neural computation hinges on the use of hardware more efficient than the conventional microprocessors. Recent works have shown that mixed-signal integrated memristive circuits, especially their passive (0T1R) variety, may increase the neuromorphic network performance dramatically, leaving far behind their digital counterparts. The major obstacle, however, is immature memristor technology so that only limited functionality has been reported. Here we demonstrate operation of one-hidden layer perceptron classifier entirely in the mixed-signal integrated hardware, comprised of two passive 20 × 20 metal-oxide memristive crossbar arrays, board-integrated with discrete conventional components. The demonstrated network, whose hardware complexity is almost 10× higher as compared to previously reported functional classifier circuits based on passive memristive crossbars, achieves classification fidelity within 3% of that obtained in simulations, when using ex-situ training. The successful demonstration was facilitated by improvements in fabrication technology of memristors, specifically by lowering variations in their I-V characteristics.