arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.16742cs.ARcs.LG

贯穿校验和:面向边缘 CNN 推理的轻量级故障检测

Carry-Through Checksum: A Lightweight Fault-Detection for CNN Inference at the Edge

  • Tallinn University of Technology(塔林理工大学)
  • University of Manchester(曼彻斯特大学)

机构由 AI 辅助整理,请以论文原文为准。

Kyrylo Nazarevych, Mohammad Hasan Ahmadilivani, Krister Kaldre, Davide Bertozzi, Jaan Raik

AI总结:

针对边缘 CNN 推理中的软错误,提出贯穿校验和方法,通过卷积层内嵌滤波器计算校验和并贯穿推理,实现单次验证的端到端检测,在 Jetson Orin NX 上以几乎零额外开销检测出 FP32/FP16 下 95.86%/86.56% 关键故障,重执行仅增 2.27% 开销。

AI中文摘要:

卷积神经网络(CNN)越来越多地部署在安全关键的边缘应用中,在这些应用中,软错误可能悄无声息地破坏推理输出并导致不安全的决策。此类应用通常依赖资源受限的嵌入式 GPU,因此需要故障检测与缓解技术,这些技术应增加极少的计算、内存和延迟开销,同时与标准 GPU 推理流水线无缝集成。现有的基于算法的容错技术依赖于矩阵增广和逐操作校验和验证,这带来了巨大的开销,对于嵌入式 GPU 上的 CNN 推理而言是难以承受的。在这项工作中,我们提出了贯穿校验和(carry-through checksum),这是一种用于嵌入式 GPU 上 CNN 推理中软错误检测的全新方案。该方法将专用的贯穿滤波器嵌入卷积层,这些滤波器从 CNN 自身的操作中计算校验和,并将其贯穿整个推理过程,从而通过单次输出验证实现端到端的错误检测。在多种 CNN 架构上的实验结果表明,所提出的方法在 FP32 和 FP16 下分别检测出 95.86% 和 86.56% 的关键故障,且几乎不增加每张图像的额外开销。检测到的故障通过重新执行来缓解,在 NVIDIA Jetson Orin NX GPU 上,整个测试集仅产生 2.27% 的运行时间开销。

英文摘要:

Convolutional Neural Networks (CNNs) are increasingly deployed in safety-critical edge applications, where soft errors can silently corrupt inference outputs and lead to unsafe decisions. Such applications typically rely on resource-constrained embedded GPUs, requiring fault detection and mitigation techniques that add minimal compute, memory, and latency overhead while integrating seamlessly with the standard GPU inference pipeline. Existing algorithm-based fault tolerance techniques rely on matrix augmentation and per-operation checksum verification, imposing substantial overhead that is prohibitive for CNN inference on embedded GPUs. In this work, we propose carry-through checksum, a fundamentally new scheme for soft-error detection in CNN inference on embedded GPUs. The method embeds dedicated carry-through filters into the convolutional layers, which compute a checksum from the CNN's own operations and propagate it through inference, enabling end-to-end error detection with a single output verification. Experimental results on multiple CNN architectures show that the proposed method detects 95.86% and 86.56% of critical faults for FP32 and FP16, respectively, at almost no additional per-image overhead. Detected faults are mitigated through re-execution, incurring only 2.27% run-time overhead across the entire test set on an NVIDIA Jetson Orin NX GPU.

补充信息

↑