TL

Tianxi Li

Assistant Professor
Unit: College of Arts and Sciences
Department: Department of Statistics
Office location and address
148 Amphitheater Way
Charlottesville, Virginia 22904
Education
Ph.D., University of Michigan, 2013-2018
M.S., Stanford University, 2010-2012
B.S., Zhejiang University, 2006-2010
Biography

My research mainly focuses on using statistical machine learning methods to extract salient pattern of data and making valid statistical inference on top of it. One major type of data I work on are networks/graphs. Such data structures are widely used in describing complex systems or social interactions. Statistical methods are able to quantify the randomness and uncertainty for such data sets and extracting key structural information of the system while ignoring the noises. I also work on statistical modeling and predictive methods on high dimensional data sets and their applications in biological and social problems. As a related area, I also have strong interests in efficient optimization methods and computing strategies on complex data sets.

Collaborative Research: Inference for Networks: Bridging the Gap between Metric Spaces and Graphs
Source: U.S. National Science Foundation (NSF)
July 01, 2020 – June 30, 2023
STAT 3280: Data Visualization and Management
Credits: 3
Introduces methods for presenting data graphically and in tabular form, including the use of software to create visualizations. Also introduced are databases, with topics including traditional relational databases and SQL (Structured Query Language) for retrieving information. Prerequisite: A prior statistics course and prior experience with coding.
STAT 5993: Directed Reading
Credits: 1–3
Research into current statistical problems under faculty supervision.
STAT 6020: Optimization and Monte Carlo Methods in Statistics and Machine Learning
Credits: 3
This course is designed to give a graduate-level student (and senior undergrads) a thorough grounding in properties about optimization and integrating problems in statistics and machine learning, and a broad comprehension of algorithms tailored to exploit such properties and some additional computational interference strategies.
STAT 6120: Linear Models
Credits: 3
Course develops fundamental methodology to regression and linear-models analysis in general. Topics include model fitting and inference, partial and sequential testing, variable selection, transformations, diagnostics for influential observations, multicollinearity, and regression in nonstandard settings. Conceptual discussion in lectures is supplemented withhands-on practice in applied data-analysis tasks using SAS or R statistical software. Prerequisite: Graduate standing in Statistics, or instructor permission.
STAT 7100: Introduction to Advanced Statistical Inference
Credits: 3
This course introduces fundamental concepts in the classical theory of statistical inference. Topics include sufficiency and related statistical principles, elementary decision theory, point estimation, hypothesis testing, likelihood-ratio tests, interval estimation, large-sample analysis, and elementary modeling applications. Prerequisite: STAT 6190 and graduate standing in Statistics
STAT 9998: Non-Topical Research, Preparation for Doctoral Research
Credits: 1–12
For doctoral research, taken before a dissertation director has been selected.
STAT 9999: Non-Topical Research
Credits: 1–12
For doctoral research, taken under the supervision of a dissertation director.
  • 2018 Honorable Mention for ProQuest Distinguished Dissertation Awards, University of Michigan.
  • 2017 Best student paper, ASA section for Nonparametric Statistics.
  • 2015 Student paper award by Monsanto, ASA section for Statistical Programmers and Analysts.
  • 2014-2015 Rackham International Student Fellowship, University of Michigan.