An Approximate Backpropagation Learning Rule for Memristor Based Neural Networks Using Synaptic Plasticity

TL;DR

This paper proposes an approximate backpropagation algorithm tailored for memristor-based neural networks, utilizing pulse-based signals and a min operation to enable efficient training compatible with memristor synapses.

Contribution

It introduces a novel approximation method for backpropagation that replaces the product with a min operation and uses pulse signals, facilitating memristor-compatible neural network training.

Findings

The approximation converges well in simulations.

It is suitable for memristor implementations.

The method simplifies backpropagation for hardware use.

Abstract

We describe an approximation to backpropagation algorithm for training deep neural networks, which is designed to work with synapses implemented with memristors. The key idea is to represent the values of both the input signal and the backpropagated delta value with a series of pulses that trigger multiple positive or negative updates of the synaptic weight, and to use the min operation instead of the product of the two signals. In computational simulations, we show that the proposed approximation to backpropagation is well converged and may be suitable for memristor implementations of multilayer neural networks.