Late Breaking Results: Energy-Efficient Printed Machine Learning Classifiers with Sequential SVMs

TL;DR

This paper presents a novel printed SVM classifier design that significantly reduces energy consumption, enabling battery-powered machine learning devices with improved efficiency and accuracy.

Contribution

It introduces sequential printed SVM circuits optimized for low power, achieving substantial energy savings over existing printed classifiers.

Findings

6.5x energy savings compared to state-of-the-art

Maintains higher accuracy with reduced energy consumption

Supports battery-powered machine learning applications

Abstract

Printed Electronics (PE) provide a mechanically flexible and cost-effective solution for machine learning (ML) circuits, compared to silicon-based technologies. However, due to large feature sizes, printed classifiers are limited by high power, area, and energy overheads, which restricts the realization of battery-powered systems. In this work, we design sequential printed bespoke Support Vector Machine (SVM) circuits that adhere to the power constraints of existing printed batteries while minimizing energy consumption, thereby boosting battery life. Our results show 6.5x energy savings while maintaining higher accuracy compared to the state of the art.