Convolutional Neural Networks (CNNs) have become integral in safety-critical
applications, thus raising concerns about their fault tolerance. Conventional
hardware-dependent fault tolerance methods, such as Triple Modular Redundancy
(TMR), are computationally expensive, imposing a remarkable overhead on CNNs.
Whereas fault tolerance techniques can be applied either at the hardware level
or at the model levels, the latter provides more flexibility without
sacrificing generality. This paper introduces a model-level hardening approach
for CNNs by integrating error correction directly into the neural networks. The
approach is hardware-agnostic and does not require any changes to the
underlying accelerator device. Analyzing the vulnerability of parameters
enables the duplication of selective filters/neurons so that their output
channels are effectively corrected with an efficient and robust correction
layer. The proposed method demonstrates fault resilience nearly equivalent to
TMR-based correction but with significantly reduced overhead. Nevertheless,
there exists an inherent overhead to the baseline CNNs. To tackle this issue, a
cost-effective parameter vulnerability based pruning technique is proposed that
outperforms the conventional pruning method, yielding smaller networks with a
negligible accuracy loss. Remarkably, the hardened pruned CNNs perform up to
24\% faster than the hardened un-pruned ones.

本文介绍了一种适用于卷积神经网络的模型级加固方法，通过将纠错直接集成到神经网络中来提高其容错性。该方法与硬件无关且不需要对底层加速器设备进行任何更改。通过分析参数的脆弱性，该方法能够复制一些选择性的滤波器 / 神经元，以便通过高效且稳健的纠错层有效地对其输出通道进行纠正。该方法与基于 TMR 的修复几乎具有相当的容错性能，但开销显著减少。此外，还提出了一种基于参数脆弱性的经济实惠的剪枝技术，优于传统的剪枝方法，能够生成更小且精度损失可忽略的网络。值得注意的是，经过加固的剪枝型卷积神经网络的性能比经过加固但未剪枝的网络快达 24%。