SpRRAM: A Predefined Sparsity Based Memristive Neuromorphic Circuit for Low Power Application

TL;DR

This paper introduces SpRRAM, a sparsity-based neuromorphic circuit that reduces power consumption and complexity in deep neural networks, verified across multiple datasets with maintained accuracy.

Contribution

It presents a predefined sparsity framework for memristive neuromorphic hardware, enhancing power efficiency and scalability over traditional fully connected networks.

Findings

Significant power reduction compared to fully connected networks

Maintained high classification accuracy across datasets

Validated on MNIST, breast cancer, and health monitoring data

Abstract

In this paper, we propose an efficient predefined structured sparsity-based ex-situ training framework for a hybrid CMOS-memristive neuromorphic hardware for deep neural network to significantly lower the power consumption and computational complexity and improve scalability. The structure is verified on a wide range of datasets including MNIST handwritten recognition, breast cancer prediction, and mobile health monitoring. The results of this study show that compared to its fully connected version, the proposed structure provides significant power reduction while maintaining high classification accuracy.