Zheng Ke
Harvard University, MA·Mathematics & Statistics
Data such as health records, social networks, rankings and large collections of text often hide simpler patterns underneath the noise. One example is the small set of topics that a large body of documents is about.
- Fields
Mathematics & Statistics
Computer Science
- Topics
statistics
machine learning
random matrix theory
- Grants
- 1 known grant, through Jul 2029
What they work on
- Use a low-dimensional simplex structure found in the spectral domain of the data to develop new methods for fitting hierarchical topic models and dynamic network models.
- Extend random matrix theory to derive sharp entry-wise error bounds and uncertainty quantification for leading eigenvectors when data noise is severely heterogeneous, and use these results to guide optimal methods.
- Generate new text datasets and use them to build language models that predict influential citations and detect flaws in financial statements.
More about the research
Finding these hidden components is hard when the data are sparse, change over time, are organized in layers, or are uneven from one part to another. This collaborative project aims to build new statistical methods and theory for uncovering such patterns and to apply them to text data and large language models (LLMs).
Why it matters
Better tools for finding hidden structure in messy text, network and relational data could help scientists and professionals in health, business and other fields draw reliable conclusions from that data. Language models that flag flaws in financial statements could support auditing and investment decisions, and citation prediction could help researchers find influential work.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Jianqing Fan
Princeton University, NJ·Mathematics & Statistics
Use a low-dimensional simplex structure found in the eigenvectors of noisy data to develop methods for fitting hierarchical topic models and dynamic network models.
Joshua Cape
University of Wisconsin-Madison, WI·Mathematics & Statistics·Early-career
Develop nonparametric spectral clustering and dimension-reduction methods that tolerate messy, mixed data, and apply them to brain network studies of schizophrenia with psychiatry collaborators.
Minge Xie
Rutgers University-New Brunswick, NJ·Mathematics & Statistics
Develop PCSI (Predictability-Computability-Stability Inference) for binary and multi-class classification, screening many algorithms by predictive performance and using bootstrap resampling to improve uncertainty estimates, robustness and accuracy.
The school
Harvard University: 42 labs on PI Pages, $1.5B a year on research, 712 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.