Haitao Mao (毛海涛)

PhD student in MSU

I work hard for easy in the future: 来小毛给他整个🔥，草！走！忽略 ጿ ኈ ቼ ዽ ጿ

I am a second-year Ph.D. student at Michigan State University supervised by Dr. Jiliang Tang. And I received my bachelor degree from the Elite Program, Software Academy at the University of Electronic Science and Technology of China in 2022. I have won the best short paper award in CIKM2021 as the first author. I am the (primary) organizer of WSDM CUP 2023, KDD CUP 2023, KDD workshop 2023, and LOG local meetup 2023 (MidNorth nUS & Beijing).

My reserch interests lies in:

  • The principle on DNN learning procedure: a novel neuron perspective:
    • Neuron Compaign Initialization strategy (Won the CIKM2021 Best Short Paper)
    • Neuron steadiness regularization (NeurIPS 2022)
    • Understanding Graph structure of Neural Network (ongoing)
  • The principle on Graph-tasks and new architecture design for learning on graph: (Theoretical inspire and Practice first)
    • Node classification: Can GNN benefits all nodes iin the same graph? (NeurIPS2023)
    • Link Prediction: What is the key factors for GNN success in Link prediction? (Under-review)
    • LLM for Graph: LLM as graph predictor, annotater, and enhancer (Under-review)
  • Find new and real-world data mining challenge (NeurIPS 2022, 2023, WSDM 2023 Oral)

My research style is (1) using toy examples, (2) building suitable assumption (3) conducting real-world experimental insights (4) deriving theoritical understanding aligning with real-world scenario. With those analyses, I can (1) understand graph-related task, datasets, model architecture design, and the learning procedure (2) find new real-world challnge and define research scenario revolving on real-world challenge. (3) Find new solution based on those insights. More details on my naive understanding on research can be found at here. Feel free to discuss with me and collaboration.

Selected publication

  • Demystifying Structural Disparity in Graph Neural Networks: Can One Size Fit All?
    Haitao Mao, Zhikai Chen, Wei Jin, Haoyu Han, Yao Ma, Tong Zhao, Neil Shah, Jiliang Tang
    collaboration with SnapChat
    NeurIPS 2023 [pdf] [code] [slides] [poster]

  • Revisiting Link Prediction: A data perspective
    Haitao Mao, Juanhui Li, Harry Shomer, Bingheng Li, Wenqi Fan, Yao Ma, Tong Zhao, Neil Shah, Jiliang Tang
    collaboration with SnapChat
    preprint [pdf] [slides]

  • Neuron Campaign for Initialization Guided by Information Bottleneck Theory
    CIKM2021 Best Short Paper
    Haitao Mao*, Lun Du*, Qiang Fu*, Xu Chen*, Shi Han, Dongmei Zhang
    [pdf] [github] [blog] [Chinese blog] [Offical Poster] [Offical Slides] [Offical Video] [AI TIME Introduction ] [AI TIME presentation Slides] [AI TIME Presentation Video] [AI TIME Report]

  • Exploring the Potential of Large Language Models (LLMs) in Learning on Graphs
    Zhikai Chen, Haitao Mao, Hang Li, Wei Jin, Hongzhi Wen, Xiaochi Wei, Shuaiqiang Wang, Dawei Yin, Wenqi Fan, Hui Liu, Jiliang Tang
    collaboration with Baidu
    preprint [pdf] [code] [slides]

Professional Experience

Visiting scholar at Hong Kong Polytechnic University (March, 2023 - September 2023)

  • supervised by Research Assistant Professor Wenqi Fan and Professor Qing Li
  • Project: Graph Neural Networks for Link Prediction

Research Intern at Baidu (March, 2022 - September, 2022)

  • Search strategy department, supervised by Lixin Zou(Now an associate professor in Wuhan University).
  • Project: Whole-page Unbiased Learning to rank (NeurIPS 2022 dataset track, WSDM CUP)

Research intern at Microsoft Research Asia (January, 2021 - November, 2021)

  • Data Knowledge Intelligent Group, supervised by Principal Researcher Qiang Fu, Researcher Lun Du, and Senior Principal Researcher Manager Shi Han.
  • Project: initialization strategy (CIKM best short paper) and regularization method (NeurIPS 2022). One news about my research.
  • Application: Source Free Unsupervised Graph Domain Adaptation
  • Co-founded MS-Intern Guitar Club with Jianan Zhao.

Research Intern in NLP lab, Nanjing University (June, 2020 - August, 2020)

  • Project: Document-level relation extraction
  • Win the best project and best performance award

Education

Michigan State University (August 2022 - present)

  • Ph.D. student in DSE lab at Michigan State University, supervised by Jiliang Tang.
  • Research topic:（1）The principle on Graph-tasks （2） new architecture design for learning on graph (3) Frontier Application on Graph

University of Electronic Science and Technology of China (September 2018 - June 2022)

Awards:

  • NeurIPS 2023 Scholar Award
  • CIKM2021 Best short paper award (first author) (1/626)
  • Excellent Student of High Education in Sichuan Province (30/763)
  • Outstanding Graduate in University of Electronic Science and Technology of China (74/763)
  • Star of tomorrow intern award in Microsoft Research Asia (top 10%)
  • National first prize in Chinese Software (20/45,000) [Github]
  • Best project and best performance award in Nanjing University, NLP lab. (3/57)
  • A+ performance (4/40) on the summer camp of the Nation University of Singapore, School of computing.

Support

This page is supported by Hanlin Lan, one of my best friends in undergraduate period. Thanks for his great help.