Welcome to Pangu Research Lab

The Pangu Lab focuses on research in many areas of software engineering and artificial intelligence, including:

  • Software engineering for artificial intelligence
  • Artificial intelligence for software engineering
  • Static and dynamic program analysis
  • Software testing and verification
  • Automatic programming

The overall goal of Pangu Research Lab is to obtain better software and artificial intelligence systems, making them more robust, reliable, and secure, and easier to specify, build, maintain, or improve.



Jianjun Zhao


Office: Room 751, West 2 Bldg
Phone: +81-92-802-3625
Email: zhao [at] ait [dot] kyushu-u [dot] ac [dot] jp
Address: 744 Motooka, Nishi-ku Fukuoka 819-0395, Japan

Yaokai Feng

Assistant Professor

Office: Room 713, West 2 Bldg
Email: fengyk [at] ait [dot] kyushu-u [dot] ac [dot] jp
Address: 744 Motooka, Nishi-ku Fukuoka 819-0395, Japan


Zeming Dong

Ph.D. student

Jinbo Du

Ph.D. student (at Shanghai Jiao Tong University)

Qiang Hu

Master student

Wentao Li

Master student

Kazuki Nakahara

Undergraduate student

Tomomi Nakamura

Master student

Hidenori Oooka

Undergraduate student

Hua Qi

Research student

Rui Rui

Master student

Jiyuan Sun

Ph.D. student

Siyuan Wang

Master student

Chao Xie

Master student

Daisuke Yamamoto

Master student

Bing Yu

Ph.D. student


Daigo Kajiwara

Undergraduate student

Haichen Wen

Master student

Yoichi Omori

Assistant Professor

Office: Room 714, West 2 Bldg
Email: yomori [at] ait [dot] kyushu-u [dot] ac [dot] jp
Address: 744 Motooka, Nishi-ku Fukuoka 819-0395, Japan

Robust Deep Learning Systems

Deep learning (DL) has achieved great success in many application domains. However, how to ensure the reliability and security of DL system remains an open problem. For example, an attacker could add adversarial perturbations often imperceptible to human eyes to an image to cause a deep neural network (DNN) to misclassify perturbed images. Traditional software represents its logic as control flows crafted by human knowledge, while a DNN characterizes its behaviors by the weights of neuron edges and the nonlinear activation functions (determined by the training data). Therefore, detecting erroneous behaviors in DNNs is different from those of traditional software in nature, which necessitates effective analysis, testing and verification approaches. We plan to take a multi-pronged approach to explore deeper understanding of defects (bugs) and adversarial examples in DL systems, and methods to guarantee the reliability and security of DL systems.

Deep Learning for Software Engineering

Deep learning can provide new capabilities and approaches for addressing software engineering problems. In this project, we will explore different software engineering activities where deep learning provides promising solutions, including software testing and debugging, program analysis and verification, software mining and analytics.

Programming with Big Code

Just like huge amounts of data on the web enabled Big Data applications, now large repositories of programs (e.g. open source code on GitHub) enable a new class of applications that leverage these repositories of "Big Code". Using Big Code means to automatically learn from existing code in order to solve software engineering tasks such as predicting software bugs, predicting program behavior, or automatically generating new code.



Room 713, West Bldg 2, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan
Graduate School of Information Science and Electrical Engineering
Kyushu University