用串行SVM电路让柔性电子分类器省电6.5倍，续航更长。
Late Breaking Results: Energy-Efficient Printed Machine Learning Classifiers with Sequential SVMs
- 设计串行定制SVM电路，适配印刷电池的功耗限制。
- 相比现有最优方案，能效提升6.5倍，准确率更高。
- 适合低功耗柔性可穿戴设备，如电池供电的智能标签。
与硅基技术相比，印刷电子（PE）为机器学习电路提供了机械柔性和低成本解决方案。然而，由于特征尺寸较大，印刷分类器受限于高功耗、大面积和高能耗，难以实现电池供电系统。本文设计了符合现有印刷电池功耗约束的串行定制支持向量机（SVM）电路，以最小化能耗，从而延长电池寿命。结果表明，在保持更高准确率的同时，相比当前最优方案实现了6.5倍的能效提升。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。