Memory System Designed for Multiply-Accumulate (MAC) Engine Based on Stochastic Computing

TL;DR

This paper introduces a novel memory system optimized for stochastic computing in CNNs, significantly reducing energy consumption and increasing efficiency compared to traditional memory architectures.

Contribution

A new memory system compatible with stochastic computing-based MAC engines for CNNs, reducing energy use and improving efficiency over conventional designs.

Findings

Energy consumption reduced by 82.1% to 0.91pJ

Energy efficiency achieved 164.8 TOPS/W

Compatible with conventional memory systems

Abstract

Convolutional neural network (CNN) achieves excellent performance on fascinating tasks such as image recognition and natural language processing at the cost of high power consumption. Stochastic computing (SC) is an attractive paradigm implemented in low power applications which performs arithmetic operations with simple logic and low hardware cost. However, conventional memory structure designed and optimized for binary computing leads to extra data conversion costs, which significantly decreases the energy efficiency. Therefore, a new memory system designed for SC-based multiply-accumulate (MAC) engine applied in CNN which is compatible with conventional memory system is proposed in this paper. As a result, the overall energy consumption of our new computing structure is 0.91pJ, which is reduced by 82.1% compared with the conventional structure, and the energy efficiency achieves 164.8 TOPS/W.