keyulu

Keyulu Xu

Email: keyulu [at] mit (dot) edu

Office: MIT Stata Center, 32-G480

Mail: 2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo 101-8430


I am a graduate student in the EECS department at MIT and a member of CSAIL and the Machine Learning group. My interests are in the theory of intelligence and reasoning.

I am fortunate to be advised by Stefanie Jegelka. Previously, I was an undergraduate at UBC, where I was fortunate to be advised by Nick Harvey. I also seasonally visit Ken-ichi Kawarabayashi at NII in Tokyo.

Recent News

Publications

Email me if you have any questions about my papers or code, or if you would like to collaborate with me.

  1. What Can Neural Networks Reason About?
    Keyulu Xu, Jingling Li, Mozhi Zhang, Simon S. Du, Ken-ichi Kawarabayashi, Stefanie Jegelka.
    Manuscript, 2019.
    [Paper] [arXiv]
  2. Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels
    Simon S. Du, Kangcheng Hou, Barnabas Poczos, Ruslan Salakhutdinov, Ruosong Wang, Keyulu Xu.
    Advances in Neural Information Processing Systems (NeurIPS) 2019.
    [Paper] [arXiv]
  3. Are Girls Neko or Shōjo?
    Cross-Lingual Alignment of Non-Isomorphic Embeddings with Iterative Normalization

    Mozhi Zhang, Keyulu Xu, Ken-ichi Kawarabayashi, Stefanie Jegelka, Jordan Boyd-Graber.
    Association for Computational Linguistics (ACL) 2019.
    [Paper] [arXiv]
  4. How Powerful are Graph Neural Networks?
    Keyulu Xu, Weihua Hu, Jure Leskovec, Stefanie Jegelka.
    International Conference on Learning Representations (ICLR) 2019. Oral Presentation.
    [Paper] [arXiv] [code]
  5. Representation Learning on Graphs with Jumping Knowledge Networks
    Keyulu Xu, Chengtao Li, Yonglong Tian, Tomohiro Sonobe, Ken-ichi Kawarabayashi, Stefanie Jegelka.
    International Conference on Machine Learning (ICML) 2018. Long talk.
    [Paper] [arXiv] [Code (soon)]
  6. Distributional Adversarial Networks
    Chengtao Li, David Alvarez-Melis, Keyulu Xu, Stefanie Jegelka, Suvrit Sra.
    International Conference on Learning Representations workshop track (ICLR) 2018.
    [arXiv]
  7. Generating Random Spanning Trees via Fast Matrix Multiplication
    Nicholas J. A. Harvey and Keyulu Xu.
    Latin American Theoretical Informatics Symposium (LATIN) 2016.
    [Paper] [Web]

Experience

I have lived and worked in some of the most exciting cities in the world -- Vancouver, Tokyo, NYC and Shanghai.

Talks

Some recent talks by Keyulu, with video if available.

How Powerful are Graph Neural Networks? at the Seventh ICLR 2019, New Orleans, Louisiana, United States.
[Slides] [Video]
Powerful Graph Neural Networks, at Kyoto University, Graduate School of Informatics, Kashima & Yamada Lab. [Slides]
Representation Learning on Graphs with Jumping Knowledge Networks, at RIKEN AIP, Nihonbashi, Tokyo. [Slides]
Representation Learning on Graphs with Jumping Knowledge Networks, at ICML 2018, Stockholm, Sweden. [Slides] [Video]
Generating Random Spanning Trees via Fast Matrix Multiplication, at LATIN 2016, Ensenada, Mexico. [Slides]