Write a Blog >>
ICSE 2022
Sun 8 - Fri 27 May 2022
Mon 9 May 2022 21:25 - 21:30 at ICSE room 2-odd hours - Reliability and Safety 4 Chair(s): Jonathan Sillito
Fri 13 May 2022 05:20 - 05:25 at ICSE room 1-odd hours - Reliability and Safety 2 Chair(s): Shahar Maoz

To analyse local robustness properties of deep neural networks (DNNs), we present a practical framework from a model learning perspective. Based on black-box model learning with scenario optimisation, we abstract the local behaviour of a DNN via an affine model with the probably approximately correct (PAC) guarantee. From the learned model, we can infer the corresponding PAC-model robustness property. The innovation of our work is the integration of model learning into PAC robustness analysis: that is, we construct a PAC guarantee on the model level instead of sample distribution, which induces a more faithful and accurate robustness evaluation. This is in contrast to existing statistical methods without model learning. We implement our method in a prototypical tool named DeepPAC. As a black-box method, DeepPAC is scalable and efficient, especially when DNNs have complex structures or high-dimensional inputs. We extensively evaluate DeepPAC, with 4 baselines (using formal verification, statistical methods, testing and adversarial attack) and 20 DNN models across 3 datasets, including MNIST, CIFAR-10, and ImageNet. It is shown that DeepPAC outperforms the state-of-the-art statistical method PROVERO, and it achieves more practical robustness analysis than the formal verification tool ERAN. Also, its results are consistent with existing DNN testing work like DeepGini.

Mon 9 May

Displayed time zone: Eastern Time (US & Canada) change

21:00 - 22:00
21:00
5m
Talk
Are We Training with The Right Data? Evaluating Collective Confidence in Training Data using Dempster Shafer Theory
NIER - New Ideas and Emerging Results
Sangeeta Dey Ajou University, Seok-Won Lee Ajou University
Pre-print Media Attached
21:05
5m
Talk
Automating Staged Rollout with Reinforcement Learning
NIER - New Ideas and Emerging Results
Shadow Pritchard University of Tulsa, Vidhyashree Nagaraju University of Tulsa, Lance Fiondella University of Massachusetts Dartmouth
Pre-print File Attached
21:10
5m
Talk
An Empirical Study on Quality Issues of eBay's Big Data SQL Analytics Platform
SEIP - Software Engineering in Practice
Feng Zhu ebay.Inc, Lijie Xu Institute of Software, Chinese Academy of Sciences, Gang Ma ebay.Inc, Shuping Ji University of Toronto, Jie Wang Peking University, China / Ant Group, China / Alibaba Group, China, Gang Wang ebay.Inc, Hongyi Zhang ebay.Inc, Kun Wan ebay.Inc, Mingming Wang ebay.Inc, Xingchao Zhang ebay.Inc, Yuming Wang ebay.Inc, Jingpin Li ebay.Inc
DOI Pre-print
21:15
5m
Talk
PerfSig: Extracting Performance Bug Signatures via Multi-modality Causal Analysis
Technical Track
Jingzhu He ShanghaiTech University, Yuhang Lin North Carolina State University, Xiaohui Gu North Carolina State University, Chin-Chia Michael Yeh Visa Research, Zhongfang Zhuang Visa Research
DOI Pre-print Media Attached
21:20
5m
Talk
TOGA: A Neural Method for Test Oracle GenerationDistinguished Paper Award
Technical Track
Elizabeth Dinella , Gabriel Ryan Columbia University, USA, Todd Mytkowicz Microsoft Research, Shuvendu K. Lahiri Microsoft Research
DOI Pre-print Media Attached
21:25
5m
Talk
Towards Practical Robustness Analysis for DNNs based on PAC-Model Learning
Technical Track
Renjue Li Institute of Software at Chinese Academy of Sciences, China, Pengfei Yang Institute of Software at Chinese Academy of Sciences, China, Cheng-Chao Huang Nanjing Institute of Software Technology, ISCAS, Youcheng Sun The University of Manchester, Bai Xue Institute of Software at Chinese Academy of Sciences, China, Lijun Zhang Institute of Software, Chinese Academy of Sciences
Pre-print Media Attached

Fri 13 May

Displayed time zone: Eastern Time (US & Canada) change

05:00 - 06:00
Reliability and Safety 2NIER - New Ideas and Emerging Results / Technical Track / Journal-First Papers at ICSE room 1-odd hours
Chair(s): Shahar Maoz Tel Aviv University, Israel
05:00
5m
Talk
TaintBench: Automatic Real-World Malware Benchmarking of Android Taint Analyses
Journal-First Papers
Linghui Luo Amazon Web Services, Felix Pauck Paderborn University, Germany, Goran Piskachev Fraunhofer IEM, Manuel Benz Paderborn University, Ivan Pashchenko University of Trento, Martin Mory Paderborn University, Eric Bodden , Ben Hermann Technical University Dortmund, Fabio Massacci University of Trento; Vrije Universiteit Amsterdam
Link to publication DOI Pre-print Media Attached File Attached
05:05
5m
Talk
Are We Training with The Right Data? Evaluating Collective Confidence in Training Data using Dempster Shafer Theory
NIER - New Ideas and Emerging Results
Sangeeta Dey Ajou University, Seok-Won Lee Ajou University
Pre-print Media Attached
05:10
5m
Talk
Multi-Intention-Aware Configuration Selection for Performance Tuning
Technical Track
Haochen He National University of Defense Technology, Zhouyang Jia National University of Defense Technology, Shanshan Li National University of Defense Technology, Yue Yu College of Computer, National University of Defense Technology, Changsha 410073, China, Chenglong Zhou National University of Defense Technology, Qing Liao Harbin Institute of Technology, Ji Wang National University of Defense Technology, Liao Xiangke National University of Defense Technology
Pre-print Media Attached
05:15
5m
Talk
PerfSig: Extracting Performance Bug Signatures via Multi-modality Causal Analysis
Technical Track
Jingzhu He ShanghaiTech University, Yuhang Lin North Carolina State University, Xiaohui Gu North Carolina State University, Chin-Chia Michael Yeh Visa Research, Zhongfang Zhuang Visa Research
DOI Pre-print Media Attached
05:20
5m
Talk
Towards Practical Robustness Analysis for DNNs based on PAC-Model Learning
Technical Track
Renjue Li Institute of Software at Chinese Academy of Sciences, China, Pengfei Yang Institute of Software at Chinese Academy of Sciences, China, Cheng-Chao Huang Nanjing Institute of Software Technology, ISCAS, Youcheng Sun The University of Manchester, Bai Xue Institute of Software at Chinese Academy of Sciences, China, Lijun Zhang Institute of Software, Chinese Academy of Sciences
Pre-print Media Attached
05:25
5m
Talk
Analyzing User Perspectives on Mobile App Privacy at Scale
Technical Track
Preksha Nema Google Inc., Pauline Anthonysamy Google Inc., Nina Taft Google Inc., Sai Teja Peddinti Google Inc.
Pre-print Media Attached

Information for Participants
Mon 9 May 2022 21:00 - 22:00 at ICSE room 2-odd hours - Reliability and Safety 4 Chair(s): Jonathan Sillito
Info for room ICSE room 2-odd hours:

Click here to go to the room on Midspace

Fri 13 May 2022 05:00 - 06:00 at ICSE room 1-odd hours - Reliability and Safety 2 Chair(s): Shahar Maoz
Info for room ICSE room 1-odd hours:

Click here to go to the room on Midspace