I am currently an associate professor at Mila-Quebec AI Institute and HEC Montreal. Prior to that, I was a Postdoc at University of Michigan and Carnegie Mellon University. I also worked at Microsoft Research Asia as an associate researcher between 2014-2016. For more information, please check my CV.

Hiring!! Our group has multiple PhD positions next Fall. In particular, we are looking for students to work on the following projects:

  • Geometric Deep Learning, Graph Neural Networks for Drug Design
  • Equivariant Neural Networks for Molecular Simulation
  • Knowledge Graph Construction and Reasoning, Natural Language Understanding

Students who are interested in working with me please apply through Mila admission (students working with me will be affiliated with UdeM) or send me an email directly.

What’s New

  • New!! Three papers are accepted to ICLR’2022. Congrats to Shengchao Liu, Meng Qu, Minkai Xu, Huiyu Cai, Chence Shi.
  • New!! Received new grant from IBM/Mila collaboration for “Pretraining molecular and protein representation with 3D structures” proposal (Oct. 1st 2021).
  • New!! Five papers are accepted to NeurIPS’2021. Congrats to Chence Shi, Zhaocheng Zhu, Louis-Pascal Xhonneux, Minghao Xu, Andreea Deac.
  • New!! Our paper “Learning interpretable cellular and gene signature embeddings from single-cell transcriptomic data”, collaborated with Prof. Yue Li at McGill Univesity, has been accepted to Nature Communications. Congrats to the leading author Yifan Zhao and Huiyu Cai.
  • New!! We just released our machine learning for drug discovery platform, TorchDrug. For more information about TorchDrug, please visit here.
  • New!! I gave an invited talk on “Geometric Deep Learning for Drug Disscovery” at ByteDance. Slides are available here.
  • New!! I will serve as an area chair at AKBC’21 (Automated Knowledge Base Construction).
  • New!! I gave a guest lecture on “Logic Reasoning for Knowledge Graphs” at the advanced machine learning class offered by Jie Tang at Tsinghua University.
  • New!! Four papers related to Drug Discovery are accepted to ICML’21. Congrats to Chence Shi, Minkai Xu, Shitong Luo, Minghao and other interns and collaborators.
  • New!! I joined a panel discussion on ‘High Impact in Practice’ at ICLR’s Geometric and Topological Representation Learning workshop.
  • New!! I gave a Guest Lecture on “Graph Representation Learning for Drug Discovery” at the Yale University.
  • New!! I gave a talk on “Graph Representation Learning for Drug Discovery” at the first workshop on “AI + Medicine” at Institute for AI Industry Research, Tsinghua University. The slides are available here
  • New!! I will serve as an area chair in NeurIPS’2021!
  • New!! I gave an invited talk “Deep generative models for molecular conformation generation” at Samsung Seminar.
  • New!! I gave an invited Talk “Graph Representation Learning for Drug Discovery” at the research seminar of School of Biomedical Informatics, University of Texas School of Biomedical Informatics. slides are available.
  • New!! I gave a keynote talk on “Learning Symbolic Logic Rules for Reasoning on Knowledge Graphs” at the International Workshop on Deep Learning on Graphs in AAAI’2021, February, 2021.Slides and video are available.
  • New!! 3 papers are accepted to ICLR’2021!!
  • New!! I introduced our recent work on “Deep Genereative Models for Molecular Conformation Genereation” at the MIT AI Cures Drug Discovery Conference, November, 2020. Slides and video are available.
  • New!! I gave a tutorial on “Neural and Symbolic Logical Reasoning on Knowledge Graphs” at the Summer School of Chinese Information Processing Society of China (CIPS), November, 2020. Slides are available here.
  • New!! I will serve as an area chair in ICML’2021!
  • New!! 3 papers are accepted to NeurIPS’2020!!
  • New!! Received a Tencent AI Lab Rhino-Bird Gift Fund. Thanks Tencent!
  • New!! 4 papers accepted to ICML’20! Congratulations to all my students and collaborators!
  • New!! Received a Amazon Faculty Research Award. Thanks Amazon!
  • New!! Received a Microsoft-Mila collaboration grant on “Towards Combining Statistical Relational Learning and Graph Neural Networks for Reasoning”. Thanks Microsoft!
  • New!! Received a Collaborative Research and Development Grant on “Intelligent Design through Graph Generation with Deep Generative Models and Reinforcement Learning” from National Research Council Canada (NRC). Thanks NRC!
  • New!! I am teaching a new course Deep Learning and Applications this semester!
  • New!! Two papers on graph representation learning for drug discovery are accepted to ICLR’2020!!
  • New!! We released the codes of the pLogicNet model in our NeurIPS’19 paper “Probabilistic Logic Neural Networks for Reasoning”
  • New!! We just released our GraphVite system, which is super effecient and only takes one minute to learn the node embeddings of a graph with one million nodes. It now supports three different tasks including node embeddings, knowledge graph embeddings, and graph&high-dimensional data visualization. For more information, check this link
  • New!! We just released a library of recommender systems with deep neural networks including session-based recommendation, feature-based recommendation, and social recommenddation. For more information, check this link
  • We just released the source codes of our RotatE model. The codes are available at link.
  • I am quite honoured to be named to the first cohort of Canada CIFAR Artificial Intelligence Chairs (CCAI Chair).CIFAR News1 CIFAR News2
  • Tutorial “Graph representation learning” by William L. Hamilton and me has been accepted by AAAI’19. See you at Hawaii!! Slides (Part 0, Part I, Part II, Part III)

Research Interests

  • Graph Representation Learning, Graph Neural Networks
  • Reasoning in Knowledge Graphs and Natural Language
  • Deep Generative Models
  • Drug Discovery
  • Material Discovery

Recent Papers

Selected Publications