PhD. Duc-Anh Nguyen

Research Interest

Teaching

Education

• 2019 - 5/2024
  Doctoral student
  VNU University of Engineering and Technology
  Topic: Methods For Ensuring The Robustness Of Machine Learning Systems
  Advisor: Pham Ngoc Hung, Associate Prof.


• 2018 - 2019
  Intership
  Department of Computer Science and Software Engineering, Auburn University, Alabama, USA
  Topic: Detect the potential trends of bug on open source projects
  Advisor: Tung Thanh Nguyen, Prof.


• 2015 - 2017
  MSc, Software Engineering
  VNU University of Engineering and Technology
  Topic: A method for test data generation for C/C++ projects
  Advisor: Pham Ngoc Hung, Associate Prof.


• 2011- 2015
  BSc, Information Technology (Computer science)
  VNU University of Engineering and Technology
  Topic: A method and tool supporting for automated testing of C programs
  Advisor: Pham Ngoc Hung, Associate Prof.

Research Experience

• 11/2022 - 10/2023
  Evaluate the robustness of deep neural networks, as project manager
  Granted by Vingroup Innovation Foundation – VINIF


• 11/2021 - 10/2022
  Evaluate the robustness of deep neural networks, as project manager
  Granted by Vingroup Innovation Foundation – VINIF


• 07/2021 - 06/2022
  Enhance the quality of deep neural networks, as project manager
  Granted by VNU University of Engineering and Technology (VNU-UET).


• 04/2021 - 09/2022
  Improve the robustness of machine learning systems, as project manager
  Granted by Toshiba Vietnam.
  We build a tool to evaluate and improve the robustness of DNN-based systems. Hence, the quality of these systems would be ensured, especially there exists perturbation on the environment.
  


• 10/2019 - 06/2020
  Automated test data generation for C/C++ projects, as leader
  Granted by FPT.
  Demo
  


• 8/2018 - 2/2019
  Bug trend prediction on Github projects
  The project aims to predict the trends of bug on Github projects over time. To detect the underlying trends of bug, we apply natural language processing (NLP) to analyze bug reports, and program analysis to analyse bug fixes. This project assists developers in understanding the potential trends of bug on their systems. Based on the predicted trend, software teams would locate the most likely erroneous modules/features, then take appropriate actions to solve the problem. I research this topic under the supervise of Prof. Tung Thanh Nguyen (Auburn University, USA).
  


• 1/2018 - 7/2018
  Parallel suggestion for C/C++ programs, as key member
  Granted by GAIO (Japan).
  The project aims to improve the quality of C/C++ programs by providing parallel suggestions for developers. Therefore, the C/C++ program would be executed more faster and effective.
  


• 2016 - 2017
  Impact prediction for web-based projects, as consultant
  When developers modify modules to fix bugs or add functions to meet the requirement of customers, it would potentially affect other modules, so bugs might arise from these changes. It is imperative to predict which modules might be changed from the change set. Therefore, my research group cooperated with Viettel, Vietnam’s largest mobile network operator, to fulfil this goal. I took responsibility for giving recommendations on architecture, design pattern, and prediction algorithm.
  


• 2016 - 2017
  Graphics user interface testing project, as leader,
  Granted by Toshiba Vietnam.
  The process of test scenarios creation is usually performed manually by testers, so it would increase the cost of the testing phase significantly. Therefore, my research group proposed a solution to address this problem in cooperation with Toshiba, a Japanese company. We would expand the limited existing test scenarios to create new ones automatically which satisfy coverage criteria.
  


• 2015 - 2016
  Change impact analysis for J2EE application, as key member
  Granted by Mitani (Japan).
  The main goal of this research is to predict the potential impact of changes in Java EE applications when developers modified modules. In this research, my main tasks include two folds: think up an effective layout algorithm for displaying a large number of components on a screen and detect the relationships between elements in source code automatically. This project is funded by Mitanni Sangyo, a well-known Japanese company.
  


• 2015 - 2016
  Test data generation for C/C++ projects, as leader
  Granted by Toshiba Vietnam.
  Demo
  Unit testing plays an important role in checking the correctness of software, so there is no denying that this phase consumes a large amount of cost. My research group aims to support testers in testing C/C++ functions. The tool would automatically generate unit tests instead of writing manually. I took responsibility for architecture, graphics user interface, and think up solutions to generate test data. Currently, the tool has been used by Toshiba and obtain much positive feedback.
  

Publication

Journal

1. Duc-Anh Nguyen, Kha Do Minh, Ngoc Nguyen Nhu, and Pham Ngoc Hung. 2023. SCADefender: An Autoencoder-Based Defense for CNN-Based Image Classifiers. In International Journal of Pattern Recognition and Artificial Intelligence (ISI)
2. Duc-Anh Nguyen, Kha Do Minh, Khoi Nguyen Le, Minh Nguyen Le, Pham Ngoc Hung. 2022. Improving Diversity and Quality of Adversarial Examples in Adversarial Transformation Network. In Soft Computing (ISI)
3. Duc-Anh Nguyen, Kha Do Minh, Pham Ngoc Hung, Nguyen Le Minh. 2022. "A Symbolic Execution-based Method to Perform Untargeted Attack on Feed-forward Neural Networks”. In Automated Software Engineering (ISI)
4. Duc-Anh Nguyen, Tran Nguyen Huong, Hieu Vo Dinh, Pham Ngoc Hung. 2019. ”On Improvements of directed automated random testing in test data generation for C++ projects”. In International Journal of Software Engineering and Knowledge Engineering (ISI)

Conference

1. Duc-Anh Pham, Duc-Anh Nguyen, Khoi Nguyen Le, and Pham Ngoc Hung. 2024. A Fast White-Box Adversarial Attack using shadows for Convolutional Neural Networks. In International Conference on knowledge and systems engineering (KSE)
2. Duc-Anh Nguyen, Duc-Manh Phan, Khoi Nguyen Le, and Pham Ngoc Hung. 2024. Generating high-quality adversarial examples for object detectors with low computational cost. In International Conference on knowledge and systems engineering (KSE)
3. Duc-Anh Nguyen, Lam Nguyen Xuan, Vu Ha Minh Trang, Pham Ngoc Hung. 2023. A Query-based Black-box Attack to Evaluate the Robustness of One-Stage Object Detectors. In International Conference on knowledge and systems engineering (KSE) (The best paper award)
4. Do Minh Kha, Duc-Anh Nguyen, Nguyen Le Minh, Pham Ngoc Hung. 2022. A Generator-based Method for Attacking Embedding-based Text Classifier. In RIVF International Conference on Computing and Communication Technologies (RIVF)
5. Doan Thi Hoai Thu, Le Duy Quang, Duc-Anh Nguyen, Pham Ngoc Hung. 2022. A Method of Automated Mock Data Generation for RESTful API Testing. In RIVF International Conference on Computing and Communication Technologies (RIVF)
6. Duc-Anh Nguyen, Kha Do Minh, Duc-Anh Pham, Pham Ngoc Hung. 2021. "Method for Improving Quality of Adversarial Examples”. In the 14th International Conference on Agents and Artificial Intelligence (ICAART - rank B)
7. Doan Thi Hoai Thu, Duc-Anh Nguyen, Pham Ngoc Hung. 2021. "Automated Test Data Generation for Typescript Web Applications”. In the 13th International Conference on knowledge and systems engineering (KSE)
8. Duc-Anh Nguyen, Do Minh Kha, Pham Thi To Nga, Pham Ngoc Hung. 2021. ”An Autoencoder-based Method for Targeted Attack on Deep Neural Network Models”. In the 15th IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF) (The best paper award)
9. Le Ba Cuong, Van Son Nguyen, Duc Anh Nguyen, Pham Ngoc Hung, and Dinh Hieu Vo. 2018. ”JCIA: A Tool for Change Impact Analysis of Java EE Applications”. In Information Systems Design and Intelligent Applications, 105–114
10. Duc Anh Nguyen and Pham Ngoc Hung. 2017. ”A Test Data Generation Method for C/C++ Projects”. In Proceedings of the Eighth International Symposium on Information and Communication Technology, 431–438
11. D. A. Nguyen, P. N. Hung, and V. H. Nguyen. 2016. ”A method for automated unit testing of C programs”. In 2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS), 17-22