Memristive Stochastic Computing for Deep Learning Parameter Optimization

TL;DR

This paper presents a novel memristive stochastic computing architecture leveraging CBRAM devices for efficient deep learning parameter optimization, significantly reducing hardware size and power consumption without sacrificing accuracy.

Contribution

It introduces a memristive stochastic computing method using CBRAM devices for deep learning optimization, achieving high efficiency and scalability.

Findings

Reduces MAC unit size by 5 orders of magnitude.

Operates with 167μW power consumption in a 40-nm CMOS process.

Maintains accuracy in CNN training for character recognition.

Abstract

Stochastic Computing (SC) is a computing paradigm that allows for the low-cost and low-power computation of various arithmetic operations using stochastic bit streams and digital logic. In contrast to conventional representation schemes used within the binary domain, the sequence of bit streams in the stochastic domain is inconsequential, and computation is usually non-deterministic. In this brief, we exploit the stochasticity during switching of probabilistic Conductive Bridging RAM (CBRAM) devices to efficiently generate stochastic bit streams in order to perform Deep Learning (DL) parameter optimization, reducing the size of Multiply and Accumulate (MAC) units by 5 orders of magnitude. We demonstrate that in using a 40-nm Complementary Metal Oxide Semiconductor (CMOS) process our scalable architecture occupies 1.55mm$^2$ and consumes approximately 167$\mu$W when optimizing parameters of a Convolutional Neural Network (CNN) while it is being trained for a character recognition task, observing no notable reduction in accuracy post-training.