论文标题
缺陷感知的设计和培训，以增强零偏头的DNN可靠性
Fault-Aware Design and Training to Enhance DNNs Reliability with Zero-Overhead
论文作者
论文摘要
深度神经网络（DNNS）可实现一系列技术进步，从临床成像到预测性工业维护和自主驾驶。但是，最近的发现表明，瞬态硬件故障可能会大大破坏模型的预测。例如，辐射引起的错误预测概率可能是如此之高，以至于妨碍DNNS模型的安全部署，敦促需要有效，有效的硬化解决方案。在这项工作中，我们建议解决培训和模型设计时间的可靠性问题。首先，我们表明香草模型受到瞬态故障的高度影响，这可能会导致表现降至37％。因此，我们基于DNN的重新设计和重新训练提供了三种零离线溶液，可以提高DNN对瞬态断层的可靠性，最高为一个数量级。我们通过广泛的消融研究对我们的工作进行补充，以量化每个硬化成分的性能的增益。
Deep Neural Networks (DNNs) enable a wide series of technological advancements, ranging from clinical imaging, to predictive industrial maintenance and autonomous driving. However, recent findings indicate that transient hardware faults may corrupt the models prediction dramatically. For instance, the radiation-induced misprediction probability can be so high to impede a safe deployment of DNNs models at scale, urging the need for efficient and effective hardening solutions. In this work, we propose to tackle the reliability issue both at training and model design time. First, we show that vanilla models are highly affected by transient faults, that can induce a performances drop up to 37%. Hence, we provide three zero-overhead solutions, based on DNN re-design and re-train, that can improve DNNs reliability to transient faults up to one order of magnitude. We complement our work with extensive ablation studies to quantify the gain in performances of each hardening component.
