Published over 60 research papers, more than 20 open-source technical reports, over 10 technical invention patents, more than 10 software copyrights, and 5+ open-source projects. Developed Think-on-Graph technology and the next-generation knowledge graph technology “Context Graph” for large language models. Proposed Guided Learning to address end-to-end deep learning training challenges in multi-stage decision-making. Co-developed the Quant4.0 and Quant5.0 paradigms for AI-driven quantitative investment, and the ideaQuant platform.
Research Experience
Returned to China in 2016 to start DeepX, developing new AI technologies for quantitative investment and risk management in financial markets; co-founded IDEA with Dr. Harry Shum in 2020. Long-term focus on AI and deep learning theory, algorithms, and applications, particularly in AI-driven quantitative investment.
Education
Graduated from the Department of Mathematical Sciences, Tsinghua University in 2003; pursued further studies at the University of Michigan, obtaining a Ph.D. in Machine Learning in 2006; served as an Assistant Professor (Tenure-track, PhD supervisor) in Computational Statistics at Harvard University from 2011.
Background
Research Interests: Quantitative Investment, Artificial Intelligence, Deep Learning. Currently, the Chief Scientist and Acting Dean of IDEA, founder of DeepX, and a concurrent professor/PhD supervisor at HKUST, Shanghai Jiao Tong University, and Tsinghua University.
Miscellany
Personal interests include sharing insights on AI and quantitative investment at various forums and conferences, such as China Fund News and Phoenix TV.