一句话：提出 DeepPerform，通过分布学习生成触发 AdNN 输入相关性能瓶颈的测试样本，高效检测性能退化。
自适应深度神经网络（AdNN）正越来越多地部署在资源受限的嵌入式设备上。论文观察到，与传统软件类似，AdNN 中存在冗余计算，会导致显著性能退化；这种退化依赖于输入，被称为输入相关性能瓶颈（IDPB）。为确保 AdNN 满足资源受限应用的性能要求，需要对其进行性能测试以检测 IDPB，但现有神经网络测试方法主要关注正确性测试，不涉及性能测试。为填补这一空白，论文提出 DeepPerform，一种可扩展的性能测试样本生成方法。作者首先将生成能检测 IDPB 的性能测试样本形式化为优化问题，然后通过学习和估计 AdNN 计算消耗分布来高效求解。评估在三个常用数据集和五个流行 AdNN 模型上进行，结果表明 DeepPerform 生成的测试样本可造成更严重性能退化（FLOPs 最高增加 552%），且生成测试输入的运行时开销仅 6–10 毫秒，显著优于基线方法。
Today, an increasing number of Adaptive Deep Neural Networks (AdNNs) are being used on resource-constrained embedded devices. We observe that, similar to traditional software, redundant computation exists in AdNNs, resulting in considerable performance degradation. The performance degradation is dependent on the input and is referred to as input-dependent performance bottlenecks (IDPBs). To ensure an AdNN satisfies the performance requirements of resource-constrained applications, it is essential to conduct performance testing to detect IDPBs in the AdNN. Existing neural network testing methods are primarily concerned with correctness testing, which does not involve performance testing. To fill this gap, we propose DeepPerform, a scalable approach to generate test samples to detect the IDPBs in AdNNs. We first demonstrate how the problem of generating performance test samples detecting IDPBs can be formulated as an optimization problem. Following that, we demonstrate how DeepPerform efficiently handles the optimization problem by learning and estimating the distribution of AdNNs’ computational consumption. We evaluate DeepPerform on three widely used datasets against five popular AdNN models. The results show that DeepPerform generates test samples that cause more severe performance degradation (FLOPs: increase up to 552%). Furthermore, DeepPerform is substantially more efficient than the baseline methods in generating test inputs (runtime overhead: only 6–10 milliseconds).
@inproceedings{paperbot7312,
title = {DeepPerform: An Efficient Approach for Performance Testing of Resource-Constrained Neural Networks},
author = {Simin Chen and Mirazul Haque and Cong Liu and Wei Yang},
booktitle = {ASE},
year = {2022},
doi = {10.1145/3551349.3561158}
}