Effect of memristor’s potentiation-depression curves peculiarities in the convergence of physical perceptrons

Abstract

In this paper, we obtain experimental potentiation-depression (P-D) curves on different manganite-based memristive systems and simulate the learning process of perceptrons for character recognition. We analyze how the specific characteristics of the P-D curves affect the convergence time -characterized by the EPOCHs-to-convergence (ETC) parameter- of the network. Our work shows that ETC is reduced for systems displaying P-D curves with relatively low granularity and non-linear and asymmetric response. In addition, we also show that noise injection during the synaptic weight actualization further reduces the ETC. The results obtained here are expected to contribute to the optimization of hardware neural networks based on memristors cross-bar arrays.

Manganite memristive devices: recent progress and emerging opportunities

“…For this, recent reports that focused this efforts on manganite-based memristors have shown significant progress in their application for neuromorphic computing and high-density memory storage. For instance, Quiñonez et al [83] explored the effect of potentiation-depression curves, which is crucial for the convergence of physical perceptrons used in neuromorphic systems. Their findings highlight the importance of understanding the dynamic behavior of these devices to optimize their performance in artificial neural networks.…”