大鹏一日同风起，扶摇直上九万里。
——李白《上李邕》
Biography
I am a first-year graduate student in Computer Science at the University of Copenhagen and currently interning in Prof. Yujun Yan’s lab. Our research focuses on graph representation learning, particularly in the context of heterophily graphs.
Before that, I obtained my Bachelor’s degree in Artificial Intelligence from Southeast University where I had the privilege of being supervised by Prof. Beilun Wang and Prof. Yuheng Jia.
During my undergraduate studies, I ranked among the top 3% and explored various research directions in the field of AI. Under the guidance of my professors, I led or participated in multiple research projects, some of which are outlined below.
Currently, my research interests include graph, representation learning, reinforcement Learning and related fields. I am actively seeking a PhD supervisor whose research aligns with my interests. If you find my profile intriguing, I would greatly appreciate the opportunity to discuss potential collaborations. Please feel free to reach out to me via email, and you can find my CV here. You can contact me at hxj557@alumni.ku.dk
Education
- 2024.09 - 2026.06, M.SC, University of Copenhagen, Computer Science
- 2020.06 - 2024.06, B.S., Southeast University, Artificial Intelligence
Honors and Awards
- 2020-2021, China National Scholarship, National Scholarship Review Committee, rank: 0.18%
- 2020-2021, Merit Student at Southeast University
Publications
Xuyuan Liu, Yinghao Cai, Qihui Yang, Yujun Yan*(* corresponding author)
(NeurlPS 2024)
Jiaqi Zhang, Yinghao Cai, Zhaoyang Wang, Beilun Wang*(* corresponding author)
(2023, Preprint) | PDF
Projects
- Identify the consistency principle in both kernel and GNN methods for graph classification tasks
- Propose a loss function that is suitable for all GNNs with layered structures
- Be responsible for the main part of the experiment
- Improve graph classification performance comprehensively on various datasets, including NCI109, IMDB-B, ogbg-molhiv and so on
The second Author
- Propose an efficient algorithm to solve the problem of high-dimensional low- rank tensor regression
- Demonstrate algorithm’s efficiency and superior performance on video classifi- cation dataset, UCF101
- Be responsible for the main part of the experiment
The second author | Paper
- Design the pipeline of knowledge graph construction
- Be responsible for critical steps including named entity recognition and algo- rithm design
- Propose an algorithm predicting prescriptions based on deep random walk
- Achieve an outstanding accuracy
Leader | Report