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

2025-01-28

AI Summary

To address the low energy efficiency, high power consumption, and poor compatibility with battery-powered flexible sensing in printed electronic systems, this paper proposes a serialized, application-specific SVM circuit architecture tailored for printing fabrication. The architecture integrates low-power analog/mixed-signal circuit design, sequential decision optimization, and device-level hardware–algorithm co-optimization to jointly enhance energy efficiency and classification accuracy under stringent power constraints imposed by printed batteries. Compared with state-of-the-art printed machine learning classifiers, the proposed design reduces energy consumption by 6.5× while achieving higher classification accuracy. It marks the first demonstration of battery-operated printed intelligent sensing with practical operational lifetime—on the order of weeks. The core innovation lies in the deep hardware-aware mapping of the SVM algorithm structure onto printable circuit flows, thereby overcoming the power bottlenecks inherent in conventional digital implementations.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Learning & Optimization for CV
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Security and privacy of machine learning and AI applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

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.

Problem

• Energy Efficiency
• Machine Learning Classifiers
• Battery-powered Systems

Innovation

• Energy-efficient ML Classifier
• Printed Sequential Support Vector Machine Circuit
• Battery-powered System Optimization