Write a Blog >>
ICSE 2022
Sun 8 - Fri 27 May 2022
Tue 10 May 2022 03:25 - 03:30 at ICSE room 2-odd hours - Program Repair 1 Chair(s): Ritu Kapur
Wed 11 May 2022 21:20 - 21:21 at ICSE room 2-odd hours - Program Repair 2 Chair(s): Hamid Bagheri

Neural networks have had discernible achievements in a wide range of applications. The wide-spread adoption also raises the concern of their dependability and reliability. Similar to traditional decision-making programs, neural networks can have defects that need to be repaired. The defects may cause unsafe behaviors, raise security concerns or unjust societal impacts. In this work, we address the problem of repairing a neural network for desirable properties such as fairness and the absence of backdoor. The goal is to construct a neural network that satisfies the property by (minimally) adjusting the given neural network’s parameters (i.e., weights). Specifically, we propose CARE (\textbf{CA}usality-based \textbf{RE}pair), a causality-based neural network repair technique that 1) performs causality-based fault localization to identify the `guilty’ neurons and 2) optimizes the parameters of the identified neurons to reduce the misbehavior. We have empirically evaluated CARE on various tasks such as backdoor removal, neural network repair for fairness and safety properties. Our experiment results show that CARE is able to repair all neural networks efficiently and effectively. For fairness repair tasks, CARE successfully improves fairness by $61.91%$ on average. For backdoor removal tasks, CARE reduces the attack success rate from over $98%$ to less than $1%$. For safety property repair tasks, CARE reduces the property violation rate to less than $1%$. Results also show that thanks to the causality-based fault localization, CARE’s repair focuses on the misbehavior and preserves the accuracy of the neural networks.

Tue 10 May

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

03:00 - 04:00
Program Repair 1Technical Track / Journal-First Papers at ICSE room 2-odd hours
Chair(s): Ritu Kapur University of Sannio
03:00
5m
Talk
Evaluating Automatic Program Repair Capabilities to Repair API Misuses
Journal-First Papers
Maria Kechagia University College London, Sergey Mechtaev University College London, Federica Sarro University College London, Mark Harman University College London
Link to publication DOI Pre-print Media Attached
03:05
5m
Talk
Improving Fault Localization and Program Repair with Deep Semantic Features and Transferred Knowledge
Technical Track
Xiangxin Meng Beihang University, Beijing, China, Xu Wang Beihang University, Hongyu Zhang University of Newcastle, Hailong Sun School of Computer Science and Engineering, Beihang University, Beijing,China, Xudong Liu Beihang University
DOI Pre-print Media Attached
03:10
5m
Talk
NPEX: Repairing Java Null Pointer Exceptions without Tests
Technical Track
Junhee Lee Korea University, South Korea, Seongjoon Hong Korea University, Hakjoo Oh Korea University
Pre-print Media Attached
03:15
5m
Talk
Neural Program Repair using Execution-based Backpropagation
Technical Track
He Ye KTH Royal Institute of Technology, Matias Martinez University of Valenciennes, Martin Monperrus KTH Royal Institute of Technology
Pre-print Media Attached
03:20
5m
Talk
Trust Enhancement Issues in Program Repair
Technical Track
Yannic Noller National University of Singapore, Ridwan Salihin Shariffdeen National University of Singapore, Xiang Gao Beihang University, China, Abhik Roychoudhury National University of Singapore
Pre-print Media Attached
03:25
5m
Talk
Causality-Based Neural Network Repair
Technical Track
Bing Sun Singapore Management University, Singapore, Jun Sun Singapore Management University, Long H. Pham Singapore University of Technology and Design, Jie Shi Huawei International
Pre-print Media Attached

Wed 11 May

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

21:00 - 22:00
Program Repair 2Technical Track / Journal-First Papers at ICSE room 2-odd hours
Chair(s): Hamid Bagheri University of Nebraska-Lincoln
21:00
5m
Talk
Learning Lenient Parsing & Typing via Indirect Supervision
Journal-First Papers
Toufique Ahmed University of California at Davis, Prem Devanbu Department of Computer Science, University of California, Davis, Vincent J. Hellendoorn Carnegie Mellon University
Link to publication DOI Pre-print Media Attached
21:05
5m
Talk
DEAR: A Novel Deep Learning-based Approach for Automated Program Repair
Technical Track
Yi Li New Jersey Institute of Technology, Shaohua Wang New Jersey Institute of Technology, Tien N. Nguyen University of Texas at Dallas
Pre-print
21:10
5m
Talk
NPEX: Repairing Java Null Pointer Exceptions without Tests
Technical Track
Junhee Lee Korea University, South Korea, Seongjoon Hong Korea University, Hakjoo Oh Korea University
Pre-print Media Attached
21:15
5m
Talk
Trust Enhancement Issues in Program Repair
Technical Track
Yannic Noller National University of Singapore, Ridwan Salihin Shariffdeen National University of Singapore, Xiang Gao Beihang University, China, Abhik Roychoudhury National University of Singapore
Pre-print Media Attached
21:20
1m
Talk
Causality-Based Neural Network Repair
Technical Track
Bing Sun Singapore Management University, Singapore, Jun Sun Singapore Management University, Long H. Pham Singapore University of Technology and Design, Jie Shi Huawei International
Pre-print Media Attached
Hide past events

Information for Participants
Info for room ICSE room 2-odd hours:

Click here to go to the room on Midspace