Yuanqi Du

Yuanqi Du is an incoming PhD student at Cornell University. He is now a researcher at a start-up company, DP Technology, working on molecule simulation. He also holds a visiting position at University of Amsterdam with Prof. Max Welling. During the past, he received Bachelor’s degree in Computer Science from George Mason University in 2021. He is fortunate to work closely Prof. Bolei Zhou at CUHK (now at UCLA), Prof. Liang Zhao at Emory University, Prof. Amarda Shehu at George Mason University. He has published several papers in top Machine Learning conferences including NeurIPS, ICML, ICLR, KDD, AAAI, etc. He is very fascinated by Sciences and very interested in developing ML tools for scientific problems, especially new knowledge discovery. He is actively collaborateing with researchers in chemistry, biology, physics, and machine learning.

I am always open to discussions about interesting ideas, drop me an email if you’d like to talk/collaborate about any of the following research topics with me.

Research Interests

  • Geometric Deep Learning
  • Deep Generative Model
  • AI for Science (Biology/Chemistry/Physics/etc)
  • Machine Learning for Discovery

News! Follow me @Twitter

  • 5/22 Our paper Equivariant Graph Neural Networks with Complete Local Frames has been accepted in ICML 2022!
  • 3/22 We have released a comprehensive survey on molecule design, MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design, welcome any comments or feedback!
  • 3/22 The sceond AI for Science workshop will be hosted together with ICML 2022 (hybrid), stay tuned for more information!
  • 3/22 We have recenlty been working on a new initiative, Learning on Graphs conference along with a stellar list of advisory board members where we aim to advance graph machine learning as a community and emphasize on the review quality! Any thoughts are welcome!
  • 12/21 Excited to announce that we (myself, Prof. Adji Bousso Dieng, Prof. Yoon Kim, Dr. Rianne van den berg, Prof. Yoshua Bengio) are hosting the second Deep Generative Models for Highly Structured Data workshop with ICLR 2022, welcome to join us (submission deadline has been extended to March 7)!
  • 10/21 GraphGT, the first Machine Learning Datasets for Graph Generation and Transformation, has been accepted in NeurIPS 2021. Welcome feedback and dataset contribution!
  • 7/21 I am proud to announce that the first AI for Science workshop will be held with NeurIPS 2021. We focus on the gaps that stifle advacenment of AI for Science this year. We invite a list of poineers in AI for Science for great talks and panels. We also introduce very exciting Attention Track, Mentorship, and Award programs. Welcome to submit your work and join us in December! Follow us on Twitter.

Selected Publications

  • MolGenSurvey: A Systematic Survey in Machine Learning Models for Molecule Design. Yuanqi Du*, Tianfan Fu*, Jimeng Sun, Shengchao Liu. Preprint at Arxiv.
  • A Survey on Deep Graph Generation: Methods and Applications. Yuanqi Du*, Yanqiao Zhu*, Yinkai Wang*, Jieyu Zhang, Qiang Liu, Shu Wu. Preprint at Arxiv.
  • Disentangled Spatiotemporal Graph Generative Models. Yuanqi Du*, Xiaojie Guo*, Hengning Cao, Yanfang Ye, Liang Zhao. In Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI) 2022. (Oral)
  • Interpreting Molecular Space with Deep Generative Models. Yuanqi Du, Xian Liu, Shengchao Liu, Jieyu Zhang, Bolei Zhou. In ELLIS ML4Molecules workshop 2021. (Oral)
  • GraphGT: Machine Learning Datasets for Graph Generation and Transformation. Yuanqi Du*, Shiyu Wang*, et al. In Neural Information Processing Systems (NeurIPS) 2021 Datasets and Benchmarks track.
  • Deep Generative Model for Spatial Networks. Xiaojie Guo*, Yuanqi Du*, Liang Zhao. In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) 2021.