Yan XIAO 肖艳
 |
Associate Professor |
About me
My current research focus is evaluating the trustworthiness of deep learning systems including trustworthiness evaluation for autonomous driving, deep neural network repair for improving the accuracy of models, input discrimination for deep learning systems. My research is within the domain of Software Engineering and Artificial Intelligence in application domains related to software bug localization, software code readability, software integration test, data mining, and big data analysis. My researches are undertaken in a broad spectrum of application domains, such as self-driving cars, deep neural network analysis, and hydrological data analysis.
I held a Research Fellow in SoC, National University of Singapore in the group of Prof. Jin Song Dong and Prof. David S. Rosenblum and under the supervision of Prof. Ivan Beschastnikh.
I received my Ph.D degree in Computer Science from City University of Hong Kong under the supervision of Prof. Jacky Keung.
Thanks to the supervision of Prof. Hong Xu and Dr. George Trimponias, I started my research career at Huawei Noah’s Ark Lab.
News
[2023-09] [New!] One paper related to NLP software testing was accepted by ASE2023.
[2023-07] [New!] One paper related to smart contract source code obfuscation was accepted by TSE.
[2023-06-14] One paper related to code smell detection was accepted by SPE.
[2023-06-01] One paper related to anomaly detection using a semi-supervised approach was accepted by IEEE Transactions on Industrial Informatics.
[2022-07-21] One paper related to distinguishing input data for deep neural networks was accepted by ASE.
[2022-07-18] One paper related to detecting anomalies and adversaries for deep neural networks was accepted by TDSC.
[2022-05-26] One survey paper related to the adversarial robustness of deep neural networks was accepted by TDSC.
[2022-05-01] One paper related to code-Comment Synchronization was accepted by TOSEM.
[2022-01-12] One paper related to predicting the precise number of software defects was accepted by IST.
[2021-04-10] One empirical paper on the impact of the distance metric and measure on SMOTE-based techniques in software defect prediction was accepted by IST.
[2021-04-10] One empirical paper on the impact of the distance metric and measure on SMOTE-based techniques in software defect prediction was accepted by IST.
[2021-06-18] One paper to alleviate the class imbalance problem in software defect prediction using complexity-based oversampling technique was accepted by IST.
[2021-05-25] One paper about a self-checking system to monitor DNN outputs and trigger an alarm if the internal layer features of the model are inconsistent with the final prediction was presented in ICSE'21.
[2021-05-13] Invited to serve as a session chair of InternetWare'20.
Work Experience
Associate Professor @ Sun Yat-sen University, Shenzhen, China, Jun. 2023 - now
Research Fellow @ National University of Singapore, Singapore, Nov. 2019 - May. 2023
Visiting Researcher @ National University of Singapore, Singapore, Jan. 2019 - Jun. 2019
Research Assistant @ Noah’s Ark Lab of Huawei, Hong Kong, Mar. 2016 - Aug. 2016
Research Assistant @ Shenzhen Research Institute of the Chinese University of Hong Kong, Shenzhen, Jul. 2015 - Oct. 2015
Talks
Self-checking deep neural networks in deployment. The 43rd International Conference on Software Engineering (ICSE) 2021. [video]
Bug localization with semantic and structural features using convolutional neural network and cascade forest. The 22nd ACM International Conference on Evaluation and Assessment in Software Engineering 2018, Christchurch, New Zealand. [slides]
Bug localization with semantic and structural features using convolutional neural network and cascade forest. CityU-CS Research Student Workshop, Best Presentation Award. [slides]
An inception architecture-based model for improving code readability classification. The 22nd ACM International Conference on Evaluation and Assessment in Software Engineering 2018, Christchurch, New Zealand. [slides]
Improving bug localization with an enhanced convolutional neural network. The 24th IEEE Asia-Pacific Software Engineering Conference 2017, Nanjing, China. [slides]
Identifying textual features of high-quality questions: an empirical study on stack overflow. The 24th IEEE Asia-Pacific Software Engineering Conference 2017, Nanjing, China. [slides]
Hydrological big data prediction based on similarity search and improved BP neural network. IEEE International Congress on Big Data 2015, Shenzhen, China.
Hydrological time series anomaly mining based on symbolization and distance measure. IEEE International Congress on Big Data 2014, Shenzhen, China.