SEMULATOR: Emulating the Dynamics of Crossbar Array-based Analog Neural System with Regression Neural Networks

TL;DR

SEMULATOR employs neural networks to accurately emulate crossbar-based analog neural systems, significantly reducing simulation time and enabling more efficient design and analysis of analog neural hardware.

Contribution

It introduces a neural network-based emulator for analog neural systems, improving accuracy and speed over traditional analytical models and circuit simulators.

Findings

Successfully emulates MAC units in analog neural systems

Reduces simulation time compared to SPICE

Demonstrates both experimental and theoretical validation

Abstract

As deep neural networks require tremendous amount of computation and memory, analog computing with emerging memory devices is a promising alternative to digital computing for edge devices. However, because of the increasing simulation time for analog computing system, it has not been explored. To overcome this issue, analytically approximated simulators are developed, but these models are inaccurate and narrow down the options for peripheral circuits for multiply-accumulate operation (MAC). In this sense, we propose a methodology, SEMULATOR (SiMULATOR by Emulating the analog computing block) which uses a deep neural network to emulate the behavior of crossbar-based analog computing system. With the proposed neural architecture, we experimentally and theoretically shows that it emulates a MAC unit for neural computation. In addition, the simulation time is incomparably reduced when it compared to the circuit simulators such as SPICE.