Write a Blog >>
ICSE 2022
Sun 8 - Fri 27 May 2022
Wed 11 May 2022 20:20 - 20:25 at ICSE room 1-even hours - Machine Learning with and for SE 7 Chair(s): Lei Ma
Thu 12 May 2022 04:25 - 04:30 at ICSE room 1-even hours - Machine Learning with and for SE 3 Chair(s): Antinisca Di Marco

Data is the fuel to models, and it is still applicable in fault localization (FL). Many existing elaborate FL techniques take the code coverage matrix and failure vector as inputs, expecting the techniques could find the correlation between program entities and failures. However, the input data is high-dimensional and extremely imbalanced since the real-world programs are large in size and the number of failing test cases is much less than that of passing test cases, which are posing severe threats to the effectiveness of FL techniques.

To overcome the limitations, we propose Aeneas, a universal data augmentation approach that gener\textbf{\underline{A}}t\textbf{\underline{e}}s sy\textbf{\underline{n}}thesized failing t\textbf{\underline{e}}st cases from reduced fe\textbf{\underline{a}}ture \textbf{\underline{s}}pace for more precise fault localization. Specifically, to improve the effectiveness of data augmentation, Aeneas applies a revised principal component analysis (PCA) first to generate reduced feature space for more concise representation of the original coverage matrix, which could also gain efficiency for data synthesis. Then, Aeneas handles the imbalanced data issue through generating synthesized failing test cases from the reduced feature space through conditional variational autoencoder (CVAE). To evaluate the effectiveness of Aeneas, we conduct large-scale experiments on 458 versions of 10 programs (from ManyBugs, SIR, and Defects4J) by six state-of-the-art FL techniques. The experimental results clearly show that Aeneas is statistically more effective than baselines, e.g., our approach can improve the six original methods by 89% on average under the Top-1 accuracy.

Wed 11 May

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

20:00 - 21:00
Machine Learning with and for SE 7SEIP - Software Engineering in Practice / Technical Track / Journal-First Papers at ICSE room 1-even hours
Chair(s): Lei Ma University of Alberta
20:00
5m
Talk
Journal First: On the Value of Oversampling for Deep Learning in Software Defect Prediction
Journal-First Papers
Rahul Yedida North Carolina State University, Tim Menzies North Carolina State University
Media Attached
20:05
5m
Talk
In-IDE Code Generation from Natural Language: Promise and Challenges
Journal-First Papers
Frank Xu Carnegie Mellon University, Bogdan Vasilescu Carnegie Mellon University, USA, Graham Neubig Carnegie Mellon University
20:10
5m
Talk
Dependency Tracking for Risk Mitigation in Machine Learning (ML) Systems
SEIP - Software Engineering in Practice
Xiwei (Sherry) Xu CSIRO Data61, Chen Wang CSIRO DATA61, Zhen Wang CSIRO Data61, Qinghua Lu CSIRO’s Data61, Liming Zhu CSIRO’s Data61; UNSW
Media Attached
20:15
5m
Talk
Strategies for Reuse and Sharing among Data Scientists in Software Teams
SEIP - Software Engineering in Practice
Will Epperson Carnegie Mellon University, April Wang University of Michigan, Robert DeLine Microsoft Research, Steven M. Drucker Microsoft Research
Pre-print Media Attached
20:20
5m
Talk
A Universal Data Augmentation Approach for Fault Localization
Technical Track
Huan Xie Chongqing University, Yan Lei School of Big Data & Software Engineering, Chongqing University, Meng Yan Chongqing University, Yue Yu College of Computer, National University of Defense Technology, Changsha 410073, China, Xin Xia Huawei Software Engineering Application Technology Lab, Xiaoguang Mao National University of Defense Technology
DOI Pre-print Media Attached
20:25
5m
Talk
Explanation-Guided Fairness Testing through Genetic Algorithm
Technical Track
Ming Fan Xi'an Jiaotong University, Wenying Wei Xi'an Jiaotong University, Wuxia Jin Xi'an Jiaotong University, Zijiang Yang Western Michigan University, Ting Liu Xi'an Jiaotong University
DOI Pre-print

Thu 12 May

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

04:00 - 05:00
Machine Learning with and for SE 3Technical Track / Journal-First Papers / SEIP - Software Engineering in Practice at ICSE room 1-even hours
Chair(s): Antinisca Di Marco University of L'Aquila
04:00
5m
Talk
In-IDE Code Generation from Natural Language: Promise and Challenges
Journal-First Papers
Frank Xu Carnegie Mellon University, Bogdan Vasilescu Carnegie Mellon University, USA, Graham Neubig Carnegie Mellon University
04:05
5m
Talk
Active Learning of Discriminative Subgraph Patterns for API Misuse Detection
Journal-First Papers
Hong Jin Kang Singapore Management University, David Lo Singapore Management University
Pre-print Media Attached
04:10
5m
Talk
Dependency Tracking for Risk Mitigation in Machine Learning (ML) Systems
SEIP - Software Engineering in Practice
Xiwei (Sherry) Xu CSIRO Data61, Chen Wang CSIRO DATA61, Zhen Wang CSIRO Data61, Qinghua Lu CSIRO’s Data61, Liming Zhu CSIRO’s Data61; UNSW
Media Attached
04:15
5m
Talk
DeepFD: Automated Fault Diagnosis and Localization for Deep Learning Programs
Technical Track
Jialun Cao Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Meiziniu LI Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Xiao Chen Huazhong University of Science and Technology, Ming Wen Huazhong University of Science and Technology, Yongqiang Tian University of Waterloo, Bo Wu MIT-IBM Watson AI Lab in Cambridge, Shing-Chi Cheung Hong Kong University of Science and Technology
DOI Pre-print Media Attached
04:20
5m
Talk
What Do They Capture? - A Structural Analysis of Pre-Trained Language Models for Source Code
Technical Track
Yao Wan Huazhong University of Science and Technology, Wei Zhao Huazhong University of Science and Technology, Hongyu Zhang University of Newcastle, Yulei Sui University of Technology Sydney, Guandong Xu University of Technology, Sydney, Hai Jin Huazhong University of Science and Technology
Pre-print Media Attached
04:25
5m
Talk
A Universal Data Augmentation Approach for Fault Localization
Technical Track
Huan Xie Chongqing University, Yan Lei School of Big Data & Software Engineering, Chongqing University, Meng Yan Chongqing University, Yue Yu College of Computer, National University of Defense Technology, Changsha 410073, China, Xin Xia Huawei Software Engineering Application Technology Lab, Xiaoguang Mao National University of Defense Technology
DOI Pre-print Media Attached
04:30
5m
Talk
DeepState: Selecting Test Suites to Enhance the Robustness of Recurrent Neural Networks
Technical Track
Zixi Liu Nanjing University, Yang Feng Nanjing University, Yining Yin Nanjing University, China, Zhenyu Chen Nanjing University
DOI Pre-print Media Attached

Information for Participants
Info for room ICSE room 1-even hours:

Click here to go to the room on Midspace