I am a research associate with the College of Computer, National University of Defense Technology (NUDT). I got my PhD at NUDT in 06/2019 and my bachelor’s degree at NUDT in 06/2013. I was visiting the Advanced Analytics Institute, University of Technology Sydney (UTS) during 09/2016-09/2018.

My research interests include machine learning, representation learning, outlier detection, mulitmodal learning and system software or optmization for DL models. I have published more than 30 high-quality research papers in these areas, most of which were published at premier data science and AI conferences or journals, like AAAI, ICML, The ACM Web Conference, IJCAI, CIKM, IEEE TKDE and Computer & Security. I have generally served as a PC member at over 10 premier international conferences including AAAI, IJCAI and a reviewer for more than 10 prestigious journals including IEEE TPAMI, IEEE TKDE, IEEE TNNLS, IPM etc.

I am always seeking highly self-motivated students for PhD, Master by Research.

Welcome to contact me!

Email: jiansonglei@nudt.edu.cn

News

• Two papers on opmizing CNN inference optimization and memory management on CPUs have been published.
• One paper on unsupervised time series anomaly detection has been accepted by TKDE.
• The paper accepted by ICASSP 2024 has be selected as one of the ICASSP 2024 Best Student Papers!
• One paper on multimodal fake news detection has been accepted by Information Processing & Management.
• One paper on log anomaly detection has been accepted by ISSRE 2023.
• One paper on semi-supervised anomaly detection has beeen accepted by Information Processing & Management.
• One paper on unsupervised anomaly detection has beeen accepted by ICML 2023.
• One paper on multimodal fake news detection has beeen accepted by ICASSP 2023.
• One paper on unsupervised network intrusion detection has been accepted by Computer & Security.
• One paper on unsupervised time series anomaly detection has release the source code.
• One paper on graph pooling has been accepted by CIKM 2022.
• One paper on graph representation has been accepted by TKDE.
• One paper on network intrusion detection system has been accepted by Computer&Security (Our first attempt of utilizing representation learning to solve the intrusion detection problem).
• One paper on few-shot learning has been accepted by CVPR 2021.
• One paper on outlier interpretation has been accepted by WWW 2021.
• One paper on representation learning for heterogeneous domains has been accepted by AAAI 2020.
• Two papers on network representation and recommender system have been accepted by AAAI 2019.
• Our paper on categorical data representation was accepted by TKDE.
• We presented tutorial on PAKDD 2018 at Melbourne, Australia.