在微型设备上用高维计算实现故障自检与根因定位
DEBUG-HD: Debugging TinyML models on-device using Hyper-Dimensional computing
- 用高维编码技术结合神经网络，提升设备端故障检测能力
- 在图像和音频数据上平均比传统方法高27%的异常检测率
- 适合无云连接、需实时调试的嵌入式安全场景
TinyML模型常在无云连接的远程动态环境中运行，易发生故障。确保其可靠性不仅需检测故障，还需定位根源。但瞬时故障、隐私问题及对安全性要求高的应用（系统不能中断调试）使得原始传感器数据难以用于离线分析。我们提出DEBUG-HD，一种专为千字节级小型设备设计的资源高效设备端调试方法，采用高维计算（HDC）。该方法引入新型HDC编码技术，利用传统神经网络，在多种图像与音频数据集上，平均比先前二值化HDC方法提升27%的输入异常检测性能。
原文摘要 · Abstract (English)
TinyML models often operate in remote, dynamic environments without cloud connectivity, making them prone to failures. Ensuring reliability in such scenarios requires not only detecting model failures but also identifying their root causes. However, transient failures, privacy concerns, and the safety-critical nature of many applications-where systems cannot be interrupted for debugging-complicate the use of raw sensor data for offline analysis. We propose DEBUG-HD, a novel, resource-efficient on-device debugging approach optimized for KB-sized tinyML devices that utilizes hyper-dimensional computing (HDC). Our method introduces a new HDC encoding technique that leverages conventional neural networks, allowing DEBUG-HD to outperform prior binary HDC methods by 27% on average in detecting input corruptions across various image and audio datasets.
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。