Assitant Prof at Tsinghua, China📖
Dr. Quanming Yao is an Assistant Professor in the Department of Electronic Engineering at Tsinghua University, where he leads a world-class research team focusing on machine learning and structural data. With over 11,000 citations and an h-index of 36, he is recognized as a global expert in automated and interpretable machine learning, pioneering contributions to graph neural networks, few-shot learning, and noise-resilient deep learning algorithms. Dr. Yao has received numerous accolades, including the Aharon Katzir Young Investigator Award, Forbes 30 Under 30, and the National Young Talents Project.
Profile
Scopus Profile
Orcid Profile
Google Scholar Profile
Education Background🎓
- Ph.D. in Computer Science and Engineering
Hong Kong University of Science and Technology (2013–2018)
Thesis: Machine Learning with a Low-Rank Regularization
Supervisor: Prof. James Kwok
- Bachelor’s in Electronic and Information Engineering
Huazhong University of Science and Technology (2009–2013)
GPA: 3.8/4.0 (Rank: 1/20)
Thesis: Large-Scale Image Classification
Supervisor: Prof. Xiang Bai
Professional Experience🌱
- Assistant Professor & Ph.D. Advisor
Tsinghua University (2021–Present)
Leads a research team in automated and interpretable machine learning for structural data.
- Senior Scientist
4Paradigm (2018–2021)
Founded and led the machine learning research team, specializing in AutoML.
- Research Intern
Microsoft Research Asia (2016–2017)
Conducted research on distributed optimization under the mentorship of Dr. Tie-Yan Liu.
Dr. Yao’s research focuses on:
- Developing scalable and interpretable automated learning methods.
- Advancing graph neural networks and AutoML to enable efficient learning from structural data.
- Designing algorithms for few-shot learning and noise-resilient training in deep neural networks.
- Bridging AI innovation with real-world applications, including drug interaction prediction and financial analytics.
Author Metrics
Publications Top Notes 📄
1. Generalizing from a Few Examples: A Survey on Few-Shot Learning
- Authors: Y. Wang, Q. Yao, J.T. Kwok, L.M. Ni
- Published in: ACM Computing Surveys
- Volume and Issue: 53(3), Pages 1–34
- Citations: 3,789 (as of 2020)
- Abstract: This survey provides a comprehensive overview of few-shot learning, exploring methods for training large deep models using limited data. It offers a roadmap for research and applications in fields requiring efficient generalization from scarce examples.
2. Co-Teaching: Robust Training Deep Neural Networks with Extremely Noisy Labels
- Authors: B. Han, Q. Yao, X. Yu, G. Niu, M. Xu, W. Hu, I. Tsang, M. Sugiyama
- Published in: Advances in Neural Information Processing Systems (NeurIPS)
- Citations: 2,539 (as of 2018)
- Abstract: This milestone paper introduces the “Co-Teaching” algorithm, which addresses challenges in training deep networks under noisy label conditions. The method demonstrates robustness and efficiency, making it a top-10 cited paper at NeurIPS 2018.
3. Meta-Graph Based Recommendation Fusion Over Heterogeneous Information Networks
- Authors: H. Zhao, Q. Yao, J. Li, Y. Song, D.L. Lee
- Published in: ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD)
- Citations: 648 (as of 2017)
- Abstract: This work develops a meta-graph-based approach for improving recommendation systems by fusing information across heterogeneous networks. It has practical implications in personalized content delivery and e-commerce applications.
4. Automated Machine Learning: From Principles to Practices
- Authors: Z. Shen, Y. Zhang, L. Wei, H. Zhao, Q. Yao
- Published in: arXiv Preprint
- Citations: 645 (as of 2018)
- Abstract: The paper outlines foundational principles and practical implementations of AutoML, highlighting its potential to democratize machine learning for diverse users and applications.
5. Non-local Meets Global: An Iterative Paradigm for Hyperspectral Image Restoration
- Authors: W. He, Q. Yao, C. Li, N. Yokoya, Q. Zhao, H. Zhang, L. Zhang
- Published in: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
- Volume and Issue: 44(4), Pages 2089–2107
- Citations: 366 (as of 2020)
- Abstract: This paper proposes an integrated framework for hyperspectral image restoration that combines non-local and global paradigms. The method significantly enhances image quality and has implications for remote sensing and environmental monitoring.
Conclusion
Dr. Quanming Yao is an exemplary candidate for the Best Researcher Award. His groundbreaking contributions to machine learning, particularly in graph neural networks and AutoML, have had a profound impact on both academia and industry. With a stellar academic record, significant citations, and prestigious awards, he stands out as a leader in his field. By enhancing industry collaborations and engaging more with public audiences, Dr. Yao can further extend the influence of his work, making him not only deserving of the award but also a role model for future researchers