将卷积神经网络部署到传感器端，用提前退出机制省电11%。
On-Sensor Convolutional Neural Networks with Early-Exits
- CNN分拆到传感器与主控协同计算，结合提前退出机制
- 在IMU上实现平均功耗4.8mA，比传统方案低11%
- 适合对能效敏感的嵌入式智能传感场景
微型机器学习（TinyML）旨在将机器学习技术集成到内存、算力和能耗受限的嵌入式设备中。近期，一种新的TinyML分支聚焦于将机器学习直接集成到传感器中，以进一步降低嵌入式设备的功耗。然而，现有文献中尚无针对直接在传感器上运行的卷积神经网络（CNN）的优化实现方案。本文首次在文献中提出在STMicroelectronics的惯性测量单元（IMU）内的智能传感器处理单元（ISPU）上，对深度优先型CNN进行优化设计与实现。该方法将CNN在ISPU与微控制器（MCU）间进行分割，并引入提前退出机制，当结果置信度足够时即停止传感器端的计算，从而显著降低功耗。在NUCLEO-F411RE开发板上，该方案实现了平均电流消耗4.8 mA，相比传统在MCU上的推理流程降低了11%，且精度保持一致。
原文摘要 · Abstract (English)
Tiny Machine Learning (TinyML) is a novel research field aiming at integrating Machine Learning (ML) within embedded devices with limited memory, computation, and energy. Recently, a new branch of TinyML has emerged, focusing on integrating ML directly into the sensors to further reduce the power consumption of embedded devices. Interestingly, despite their state-of-the-art performance in many tasks, none of the current solutions in the literature aims to optimize the implementation of Convolutional Neural Networks (CNNs) operating directly into sensors. In this paper, we introduce for the first time in the literature the optimized design and implementation of Depth-First CNNs operating on the Intelligent Sensor Processing Unit (ISPU) within an Inertial Measurement Unit (IMU) by STMicroelectronics. Our approach partitions the CNN between the ISPU and the microcontroller (MCU) and employs an Early-Exit mechanism to stop the computations on the IMU when enough confidence about the results is achieved, hence significantly reducing power consumption. When using a NUCLEO-F411RE board, this solution achieved an average current consumption of 4.8 mA, marking an 11% reduction compared to the regular inference pipeline on the MCU, while having equal accuracy.
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。