Assistant Professor, Department of Electronic Engineering, Tsinghua University, China
Short Biography (150 words): Quanming Yao (姚权铭) is an Assistant Professor in the Department of Electronic Engineering at Tsinghua University. A renowned machine learning researcher, Yao has focused on parsimonious deep learning, driving innovation by using knowledge-driven approaches rather than scaling laws. He developed automated graph learning methods that secured 1st place in the Open Graph Benchmark and led to the commercialization of these methods by AI unicorn 4Paradigm. His work on Drug-Drug Interaction (DDI) prediction in Nature Computational Science revolutionized the field with interpretable deep learning models. Yao is the co-founder of Kongfoo Technology, a synthetic biology startup. He has received several prestigious awards, including the Forbes 30 Under 30 and Google Fellowship. With over 100 publications and an h-index of 36, Yao is recognized globally for his contributions to machine learning and AI research.
Pofile
Education
Quanming Yao holds a Ph.D. in Computer Science and Engineering from the Hong Kong University of Science and Technology (HKUST), which he completed between September 2013 and June 2018. His academic journey began with a Bachelor’s degree in Electronic and Information Engineering from HuaZhong University of Science and Technology (HZUST), where he studied from September 2009 to June 2013. Yao’s academic training laid the foundation for his groundbreaking research in machine learning, particularly in parsimonious deep learning and graph learning. His research during his doctoral studies focused on machine learning theory and its practical applications, culminating in innovative methods that have since impacted both academia and industry. His doctoral research was recognized with prestigious awards, including the Google Ph.D. Fellowship in 2016. Yao’s commitment to excellence in his studies has contributed significantly to his reputation as a rising star in the field of AI and machine learning.
Experience
Quanming Yao’s professional experience spans both academia and industry. Since 2021, he has served as an Assistant Professor at Tsinghua University’s Department of Electronic Engineering, where he also mentors Ph.D. students. Before joining Tsinghua, Yao worked as a Senior Scientist at 4Paradigm Inc., a leading AI company based in Hong Kong, from June 2018 to May 2021. At 4Paradigm, Yao was involved in research and product development, specifically developing automated graph learning techniques that have been commercialized in the company’s products. Yao’s expertise extends to multiple domains, including drug discovery, where his work on Drug-Drug Interaction prediction has led to a new approach in biomedical research. Additionally, Yao co-founded Kongfoo Technology, a synthetic biology startup, demonstrating his ability to apply AI in real-world applications. His work is frequently cited in leading journals and conferences, making him a significant contributor to machine learning research globally.
Awards and Honors
Quanming Yao’s outstanding contributions to machine learning have earned him numerous prestigious awards. In 2024, he received the inaugural Intech Prize from Ant Group, recognizing him as one of the most outstanding young scholars in computer science. He was also named in the Forbes 30 Under 30 list for Science & Healthcare in China in 2020, highlighting his achievements in AI and technology. Yao has been recognized as a “Global Top Chinese New Star in Machine Learning” since 2022 and has received the Aharon Katzir Young Investigator Award in 2023. He was also selected for the National Young Talents Project in 2020, a high-level recognition for young scientists in China. Other accolades include the Wuwen Jun Prize for Excellence in AI, the Google Ph.D. Fellowship in Machine Learning, and the Young Scientist Award in Hong Kong. These honors reflect Yao’s remarkable impact in AI research and innovation.
Research Focus
Quanming Yao’s research primarily focuses on parsimonious deep learning, where he aims to achieve impressive results with minimal complexity. His work emphasizes knowledge-driven solutions, challenging traditional scaling laws that often drive deep learning innovation. Yao has made significant strides in developing automated graph learning methods, which have been highly successful, including securing 1st place in the Open Graph Benchmark. His groundbreaking work on interpretable Drug-Drug Interaction (DDI) prediction, as published in Nature Computational Science, stands as a prime example of his approach to making deep learning methods more accessible and applicable to real-world problems, such as drug discovery. Yao also explores novel methods like low-rank tensor learning with nonconvex regularization, improving the speed and efficiency of machine learning optimization processes. His research has wide-ranging implications, from graph learning and drug interaction prediction to broader AI applications, showcasing his ability to bridge theoretical and practical advancements in AI.
Publications
- Emerging Drug Interaction Prediction Enabled by Flow-based Graph Neural Network with Biomedical Network 🧬 (Nature Computational Science, 2023)
- AutoBLM: Bilinear Scoring Function Search for Knowledge Graph Learning 📊 (IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022)
- Efficient Low-rank Tensor Learning with Nonconvex Regularization 🔢 (Journal of Machine Learning Research, 2022)
- Efficient Learning with Nonconvex Regularizers by Nonconvexity Redistribution 🔍 (Journal of Machine Learning Research, 2018)
- Co-teaching: Robust Training Deep Neural Networks with Extremely Noisy Labels 🏷️ (NeurIPS, 2018)