Toward understanding deep learning framework bugs

J Chen, Y Liang, Q Shen, J Jiang, S Li - ACM Transactions on Software …, 2023 - dl.acm.org
DL frameworks are the basis of constructing all DL programs and models, and thus their
bugs could lead to the unexpected behaviors of any DL program or model relying on them …

Toward improving the robustness of deep learning models via model transformation

Y Zhang, Z Wang, J Jiang, H You, J Chen - Proceedings of the 37th IEEE …, 2022 - dl.acm.org
Deep learning (DL) techniques have attracted much attention in recent years, and have
been applied to many application scenarios, including those that are safety-critical …

Assessing the reliability of deep learning classifiers through robustness evaluation and operational profiles

X Zhao, W Huang, A Banks, V Cox, D Flynn… - arXiv preprint arXiv …, 2021 - arxiv.org
The utilisation of Deep Learning (DL) is advancing into increasingly more sophisticated
applications. While it shows great potential to provide transformational capabilities, DL also …

Attack as detection: Using adversarial attack methods to detect abnormal examples

Z Zhao, G Chen, T Liu, T Li, F Song, J Wang… - ACM Transactions on …, 2024 - dl.acm.org
As a new programming paradigm, deep learning (DL) has achieved impressive performance
in areas such as image processing and speech recognition, and has expanded its …

Reliability assessment and safety arguments for machine learning components in system assurance

Y Dong, W Huang, V Bharti, V Cox, A Banks… - ACM Transactions on …, 2023 - dl.acm.org
The increasing use of Machine Learning (ML) components embedded in autonomous
systems—so-called Learning-Enabled Systems (LESs)—has resulted in the pressing need …

CrossCert: A Cross-Checking Detection Approach to Patch Robustness Certification for Deep Learning Models

Q Zhou, Z Wei, H Wang, B Jiang, WK Chan - Proceedings of the ACM on …, 2024 - dl.acm.org
Patch robustness certification is an emerging kind of defense technique against adversarial
patch attacks with provable guarantees. There are two research lines: certified recovery and …

Predictive mutation analysis of test case prioritization for deep neural networks

Z Wei, H Wang, I Ashraf… - 2022 IEEE 22nd …, 2022 - ieeexplore.ieee.org
Testing deep neural networks requires high-quality test cases, but using new test cases
would incur the labor-intensive test case labeling issue in the test oracle problem. Test case …

A Post-training Framework for Improving the Performance of Deep Learning Models via Model Transformation

J Jiang, J Yang, Y Zhang, Z Wang, H You… - ACM Transactions on …, 2024 - dl.acm.org
Deep learning (DL) techniques have attracted much attention in recent years and have been
applied to many application scenarios. To improve the performance of DL models regarding …

A two-stage framework for ambiguous classification in software engineering

J Li, Y Lei, S Li, H Zhou, Y Yu, Z Jia… - 2023 IEEE 34th …, 2023 - ieeexplore.ieee.org
Classification tasks are prevalent and play a crucial role in the field of software engineering.
However, when two classes exhibit similar features at the class level, the classification …

DeepFeature: Guiding adversarial testing for deep neural network systems using robust features

L Feng, X Wang, S Zhang, Z Zhao - Journal of Systems and Software, 2025 - Elsevier
With the deployment of Deep Neural Network (DNN) systems in security-critical fields, more
and more researchers are concerned about DNN robustness. Unfortunately, DNNs are …