Methodology for Realizing VMM with Binary RRAM Arrays: Experimental Demonstration of Binarized-ADALINE Using OxRAM Crossbar

TL;DR

This paper introduces a hardware methodology for implementing vector matrix multiplication on RRAM arrays, demonstrating a binarized-ADALINE classifier with experimental results showing promising accuracy and highlighting practical challenges.

Contribution

It presents a novel hardware mapping technique for VMM on RRAM arrays and experimentally demonstrates a binarized-ADALINE classifier using an OxRAM crossbar.

Findings

Achieved 78% classification accuracy in simulation.

Achieved 67% classification accuracy in experiments.

Identified device variability and sneak-path issues as main accuracy limitations.

Abstract

In this paper, we present an efficient hardware mapping methodology for realizing vector matrix multiplication (VMM) on resistive memory (RRAM) arrays. Using the proposed VMM computation technique, we experimentally demonstrate a binarized-ADALINE (Adaptive Linear) classifier on an OxRAM crossbar. An 8x8 OxRAM crossbar with Ni/3-nm HfO2/7 nm Al-doped-TiO2/TiN device stack is used. Weight training for the binarized-ADALINE classifier is performed ex-situ on UCI cancer dataset. Post weight generation the OxRAM array is carefully programmed to binary weight-states using the proposed weight mapping technique on a custom-built testbench. Our VMM powered binarized-ADALINE network achieves a classification accuracy of 78% in simulation and 67% in experiments. Experimental accuracy was found to drop mainly due to crossbar inherent sneak-path issues and RRAM device programming variability.