Hi here, I'm Linghao CHEN (陈凌灏 in Chinese)! Now, I am a student in School of Computer Science and Technology at Xidian University. My major is software engineering. I am
expected to graduate in 2022. My research interest includes Graph representation learning, Anomaly detection and Computer Vision. I have worked as a Research Intern in International Digital Economy Academy (IDEA) since Nov. 2021. Now, I am also a student member of IEEE.
Address: Building #7, No.3 Begonia College(海棠3号书院), 266 Xinglong Section of Xifeng Road, Xi'an,
ShaanXi, P.R. China.
Motto: Seek the truth, practice real skills and do real things! (求真学问，练真本领，做真东西！)
2018 ~ 2019, National Scholarship(<1%), School-level outstanding student
Sep. 2020, The National College Student Mathematical Modeling Competition (National
level), First Prize(<0.7%) [details].
Apr. 2021, MCM/ICM, Meritorious Winner.
Apr. 2020, MCM/ICM, Honorable Mention.
Sep. 2019, The National College Student Mathematical Modeling Competition (Shaanxi Division),
Sep. 2019, The 12th College Student Mathematics Competition in Shaanxi Province, First
Oct. 2020, The 12th National College Student Mathematics Competition, Second
Oct. 2019, The 11th National College Student Mathematics Competition, Second
Introduction to My Research Interests
Graph Representation Learning (Especially GNNs)
With the development of deep learning, people try to use vector, matrix or other forms to represent the nodes in a graph.
The representation learning of graph is mainly composed of Graph Embedding and Graph Neural Networks (GNNs) and I mainly focus on GNNs.
The mainstream technologies of GNNs can be divided into spectral methods and spatial methods.
Although most spatial methods seem to be an easier way to represent, I think the improvement of GNNs' expressive ability depends on spectral methods.
Most people concentrate on the low frequency signals in graph filters, but rare notice that high frequency signals are also meaningful to us.
There are still many problems need us to solve, like over-smoothing, designing a "deep" GNN, improving its expressive ability and so on.
Anomaly Detection on Graph
Anomaly Detection developed rapidly in past decades, but graph-based anomaly detection technologies have only attracted attention in recent years.
In graph-based anomaly detection, I think GNNs can have a lot of room to develop.
As a non-linear high-dimensional function, GNNs do not rely on assumptions of anomalous data distribution (such as Gaussian distribution, outlier assumption, etc.),
but act as low-pass filters to effectively filter the high-frequency components in the anomalous data.
Is it a good thing that GNNs filter out anomalous high-frequency signals?
I think the answer is "YES" and my latest work (now in review) is to answer this question.
Besides, there are also many problems for us to solve about this task.
DetectorNet: Transformer-enhanced Spatial Temporal Graph Neural Network for Traffic Prediction.
Co-author with He Li, Shiyu Zhang, Xuejiao Li, Hongjie Huang, Liangcai Su, Duo Jin, Jianbin Huang and Jaesoo Yoo.
In International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL) 2021.
Rank: 1/398 (less than 1%)
CET4 / CET6: PASSED
I has been the chairman of the Inspur Student Club(ISC) [HomePage]
of Xidian University from Sep. 2020 to Sep. 2021. I am also mainly responsible for the Machine Learning and Data Mining
Group in the club.
Nov. 2021. I work as a Research Intern in International Digital Economy Academy (IDEA).
Nov. 2021. I was awarded the Fisrt-class CASA Scholarship of China Aerospace Science and Technology Corporation (the ONLY student in Xidian University).
Sep. 2021. I was awarded the National Scholarship 2020-2021.
Sep. 2021. I am selected as the Ph.D. candidate of SIGS@THU.
Aug. 2021. One paper "DetectorNet: Transformer-enhanced Spatial Temporal Graph Neural Network for Traffic Prediction" is accepted by International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL) 2021.
July 2021. A College Student Innovation and Entrepreneurship Training Program (National level) is qualified. Thanks our advisor Hong Han, group Leader Zhendong Jin and our collaborators.
June 2021. Linghao CHEN became a member of the Communist Party of China.
W.H. Yang, Ph. D. candidate of LAMDA@NJU, focus on Machine Learning (especially Online Learning) and Computer Vision.
Y.R. Pang, M.Phil. candidate of TANK LAB@TJU, focus on Web development and Computer Networks.
X.Y. Sun (sund), postgraduate of XJTU, focus on Air-gapped Attack, Covert Communication and Wireless Sensing.
B.Y. Sun (BB Chan), postgraduate of NKU, focus on Computer Vision.
M. Chen, postgraduate of ShanghaiTech, focus on Artificial Intelligence and Reinforcement Learning.
C.Z. Ran, undergraduate of XDU, focus on Artificial Intelligence and Computer Vision.
D.C. Chen (Birdie), undergraduate of XDU, focus on Algorithm Analysis and Design, chair of ISC@XDU.
Y.J. Zhang (YjmStr), undergraduate of XDU, focus on Algorithm Analysis and Design, mentor of ACM group in ISC@XDU.
Y.L. Feng, undergraduate of XDU, focus on JAVA development and Big Data.
Y.K. Xu (Viking), undergraduate of XDU, focus on Computer Security, mentor of CTF group in ISC@XDU.