[论文解读] Black-Box Testing of Deep Neural Networks Through Test Case Diversity

Zohreh Aghababaeyan, Manel Abdellatif+3 more|arXiv (Cornell University)|Dec 20, 2021

Adversarial Robustness in Machine Learning被引用 4

一句话总结

本文提出几何多样性作为黑盒、输入级别的度量指标，用于指导深度神经网络（DNN）测试，无需访问模型内部结构或训练数据。结果表明，几何多样性在预测故障检测能力方面优于最先进的白盒覆盖度量指标，且计算成本更低，使其成为基于图像的DNN测试中测试充分性更有效且实用的替代指标。

ABSTRACT

Deep Neural Networks (DNNs) have been extensively used in many areas including image processing, medical diagnostics, and autonomous driving. However, DNNs can exhibit erroneous behaviours that may lead to critical errors, especially when used in safety-critical systems. Inspired by testing techniques for traditional software systems, researchers have proposed neuron coverage criteria, as an analogy to source code coverage, to guide the testing of DNN models. Despite very active research on DNN coverage, several recent studies have questioned the usefulness of such criteria in guiding DNN testing. Further, from a practical standpoint, these criteria are white-box as they require access to the internals or training data of DNN models, which is in many contexts not feasible or convenient. In this paper, we investigate black-box input diversity metrics as an alternative to white-box coverage criteria. To this end, we first select and adapt three diversity metrics and study, in a controlled manner, their capacity to measure actual diversity in input sets. We then analyse their statistical association with fault detection using four datasets and five DNN models. We further compare diversity with state-of-the-art white-box coverage criteria. Our experiments show that relying on the diversity of image features embedded in test input sets is a more reliable indicator than coverage criteria to effectively guide the testing of DNNs. Indeed, we found that one of our selected black-box diversity metrics far outperforms existing coverage criteria in terms of fault-revealing capability and computational time. Results also confirm the suspicions that state-of-the-art coverage metrics are not adequate to guide the construction of test input sets to detect as many faults as possible with natural inputs.

研究动机与目标

• 为解决白盒覆盖度量指标在DNN测试中的局限性，这些指标需要访问模型内部结构，且在使用自然输入时与故障检测的相关性较弱。
• 开发一种仅依赖输入数据多样性的黑盒测试方法，避免对被测DNN的执行或输出进行分析。
• 评估输入多样性度量指标是否可作为DNN中故障检测能力的可靠替代指标。
• 在故障检测能力和计算效率方面，对比多样性度量指标与最先进的覆盖度量指标的性能。
• 提出一种基于聚类的方法来估计故障，以考虑由于相似故障模式导致的冗余误判。

提出的方法

• 在基于图像的DNN测试数据集上，适配并评估了三种输入多样性度量指标——几何多样性、基于特征的多样性以及基于NCD的多样性。
• 采用受控实验设置，使用四个数据集和五个预训练的DNN模型，评估多样性与故障检测之间的关系。
• 提出一种基于聚类的故障估计技术，通过按相似性对误判输入进行分组，减少冗余故障的重复计数。
• 在多种DNN架构和数据集上，测量多样性度量指标与故障检测率之间的统计关联性。
• 将最佳多样性度量指标的性能与最先进的覆盖度量指标（如神经元覆盖度）在故障检测相关性和执行时间方面进行对比。
• 选取并复现先前工作的结果，以确保与现有基于覆盖度量方法的公平比较。

实验结果

研究问题

• RQ1 黑盒输入多样性度量指标是否可在不访问模型内部结构的情况下，有效替代DNN中的故障检测？
• RQ2 在使用自然测试输入时，几何多样性与最先进的白盒覆盖度量指标相比，在预测故障检测方面表现如何？
• RQ3 输入多样性与实际故障检测的相关性如何？其在计算效率方面是否优于覆盖度量？
• RQ4 基于聚类的方法是否可通过识别由于相似故障模式导致的冗余误判，从而改善故障估计？
• RQ5 输入多样性与DNN测试套件中检测到的不同故障数量之间是否存在显著的统计关联？

主要发现

• 几何多样性作为最有效的黑盒多样性度量指标，显著优于现有的白盒覆盖度量指标，在预测故障检测方面表现更优。
• 在所有数据集和模型中，最佳多样性度量指标与故障检测之间的统计相关性强于所有评估的覆盖度量指标。
• 与基于覆盖度的方法相比，几何多样性在实现更高故障检测能力的同时，计算成本显著更低。
• 研究证实，最先进的覆盖度量指标在指导自然测试输入选择以实现有效故障检测方面作用有限。
• 基于聚类的故障估计方法成功通过识别相似误判组，减少了故障的重复计数，提高了故障检测测量的准确性。
• 结果表明，与覆盖度量相比，输入多样性（尤其是几何多样性）作为黑盒DNN测试的度量指标，在现实场景中更具可靠性与实用性。