Program slicing has been widely applied in a variety of software engineering tasks. However, existing program slicing techniques only deal with traditional programs that are constructed with instructions and variables, rather than neural networks that are composed of neurons and synapses.
In this paper, we introduce NNSlicer, the first approach for slicing deep neural networks based on data-flow analysis. Our method understands the reaction of each neuron to an input based on the difference between its behavior activated by the input and the average behavior over the whole dataset. Then we quantify the neuron contributions to the slicing criterion by recursively backtracking from the output neurons, and calculate the slice as the neurons and the synapses with larger contributions.
We demonstrate the usefulness and effectiveness of NNSlicer with three applications, including adversarial input detection, model pruning, and selective model protection. In all applications,
NNSlicer significantly outperforms other baselines that do not rely on data flow analysis.
Tue 10 Nov Times are displayed in time zone: (UTC) Coordinated Universal Time change
01:30 - 02:00
|Correlations between Deep Neural Network Model Coverage Criteria and Model Quality|
Shenao YanRutgers University, USA, Guanhong TaoPurdue University, USA, Xuwei LiuPurdue University, USA, Juan ZhaiRutgers University, USA, Shiqing MaRutgers University, USA, Lei XuNanjing University, China, Xiangyu ZhangPurdue UniversityDOI
|Deep Learning Library Testing via Effective Model GenerationACM SIGSOFT Distinguished Paper Award|
Zan WangTianjin University, China, Ming YanTianjin University, China, Junjie ChenTianjin University, China, Shuang LiuTianjin University, China, Dongdi ZhangTianjin University, ChinaDOI
|Detecting Numerical Bugs in Neural Network ArchitecturesACM SIGSOFT Distinguished Paper Award|
Yuhao ZhangPeking University, Luyao RenPeking University, China, Liqian ChenNational University of Defense Technology, China, Yingfei XiongPeking University, Shing-Chi CheungHong Kong University of Science and Technology, China, Tao XiePeking UniversityDOI
|Dynamic Slicing for Deep Neural Networks|
Ziqi ZhangPeking University, China, Yuanchun LiMicrosoft Research, China, Yao GuoPeking University, Xiangqun ChenPeking University, Yunxin LiuMicrosoft Research, ChinaDOI
|Grammar Based Directed Testing of Machine Learning Systems|
|Is Neuron Coverage a Meaningful Measure for Testing Deep Neural Networks?|
Fabrice Harel-CanadaUniversity of California at Los Angeles, USA, Lingxiao WangUniversity of California at Los Angeles, USA, Muhammad Ali GulzarUniversity of California at Los Angeles, USA, Quanquan GuUniversity of California at Los Angeles, USA, Miryung KimUniversity of California at Los Angeles, USADOI
|Operational Calibration: Debugging Confidence Errors for DNNs in the Field|
Zenan LiNanjing University, China, Xiaoxing MaNanjing University, China, Chang XuNanjing University, China, Jingwei XuNanjing University, China, Chun CaoNanjing University, China, Jian LuNanjing University, ChinaDOI
|Conversations on ML Testing 1|
Fabrice Harel-CanadaUniversity of California at Los Angeles, USA, Ming YanTianjin University, China, Sakshi UdeshiSingapore University of Technology and Design, Shenao YanRutgers University, USA, Yuhao ZhangPeking University, Zenan LiNanjing University, China, Ziqi ZhangPeking University, China, M: Hamid BagheriUniversity of Nebraska-Lincoln, USA