一句话：提出将 RNN 抽象为状态转移系统，并设计相似度度量与覆盖准则，用于对抗样本检测和覆盖引导测试生成。
针对前馈深度学习系统分析已有进展、但 RNN 等有状态深度学习系统缺乏定量分析的问题，本文首次提出面向 RNN 的定量分析框架 DeepStellar。它将 RNN 建模为抽象状态转移系统以刻画内部行为，并在此基础上设计两种轨迹相似度度量和五种覆盖准则。进一步提出基于这些度量的对抗样本检测与覆盖引导测试生成算法。作者在四个基于 RNN 的系统上评估，覆盖图像分类与自动语音识别。结果表明，抽象模型能有效捕捉 RNN 内部行为；相似度度量可区分带有极小扰动的样本，对抗样本检测准确率达 97%；覆盖准则有助于揭示错误行为，生成的对抗样本比随机测试多三倍，比展开方法多数百倍。
Deep Learning (DL) has achieved tremendous success in many cutting-edge applications. However, the state-of-the-art DL systems still suffer from quality issues. While some recent progress has been made on the analysis of feed-forward DL systems, little study has been done on the Recurrent Neural Network (RNN)-based stateful DL systems, which are widely used in audio, natural languages and video processing, etc. In this paper, we initiate the very first step towards the quantitative analysis of RNN-based DL systems. We model RNN as an abstract state transition system to characterize its internal behaviors. Based on the abstract model, we design two trace similarity metrics and five coverage criteria which enable the quantitative analysis of RNNs. We further propose two algorithms powered by the quantitative measures for adversarial sample detection and coverage-guided test generation. We evaluate DeepStellar on four RNN-based systems covering image classification and automated speech recognition. The results demonstrate that the abstract model is useful in capturing the internal behaviors of RNNs, and confirm that (1) the similarity metrics could effectively capture the differences between samples even with very small perturbations (achieving 97% accuracy for detecting adversarial samples) and (2) the coverage criteria are useful in revealing erroneous behaviors (generating three times more adversarial samples than random testing and hundreds times more than the unrolling approach).
@inproceedings{paperbot5092,
title = {DeepStellar: model-based quantitative analysis of stateful deep learning systems},
author = {Xiaoning Du and Xiaofei Xie and Yi Li and Lei Ma and Yang Liu and Jianjun Zhao},
booktitle = {FSE},
year = {2019},
doi = {10.1145/3338906.3338954}
}