Tang's interview is the first of our Faculty Spotlight series, which will introduce you to the people who make up the Department of Biostatistics and contribute to its continued success.
Lu Tang joined the Department of Biostatistics as an assistant professor on August 1, 2018. He received his PhD in biostatistics from the University of Michigan. He is developing an outstanding research program in statistical machine learning and methods for modern high dimensional data. These are extremely important areas for the department as we build for the future.
Q: Can you tell us about yourself?
I did two years of my undergraduate work in computational mathematics in China before transferring to the University of Virginia, where I completed my bachelor's training in mathematics with a minor in computer science. During that time, I took some courses in the statistics department and quickly developed interest in the field, especially machine learning. So I also stayed for my master’s training in statistics. After that, I went to the University of Michigan for my PhD training in biostatistics. It was a wonderful five years that I worked with Peter Song, a professor of biostatistics, on my dissertation topic: statistical methods of data integration, model fusion, and heterogeneity detection in big biomedical data analysis. Most progress on my thesis was made during cold Michigan winters, when I had every incentive to stay in my office and work. Ann Arbor was also the place where I met my wife and had our son.
Q: What brought you to the University of Pittsburgh?
The driving factor that brought me here was the research environment. The Department of Biostatistics provides great support for me to continue working on the research programs pertaining to my area of expertise in statistical learning and high-dimensional data analysis. The diverse and strong faculty group here also facilitates methodological collaboration among statisticians and biostatisticians. Outside of the department, there are a tremendous amount of collaboration opportunities related to health sciences on campus, where biostatisticians are in high demand. In general, the research setup here is very similar to Michigan Biostatistics, so I already feel comfortable being here on day one. Aside from research, I am also very excited about the opportunities to teach, as sharing knowledge is a very satisfying experience for me.
Q: Tell us about your research interests and why you are passionate about this topic?
I am interested in statistical methods for data integration, a relatively new research area that studies the combination of data and knowledge from different sources in order to achieve better statistical performance. The forms of data and knowledge being combined can be highly inconsistent, which call for novel statistical methods to handle the heterogeneity. I am also interested in developing new tools for analyzing complicated high-dimensional data, and applying them to solve problems in bioinformatics, clinical trials, environmental health sciences, and nutritional sciences. The reason why these research topics excite me the most is that every problem can be very different, so I feel I am not constrained to one specific domain.
Q: What is some of the best advice you’ve received?
A good one I received lately is by Professor Xiao-Li Meng from Harvard Statistics at a conference on statistical foundations. He said that young generation statisticians should take time to read publications from the early times. Many of them were very rich in ideas, for example, the works by Fisher.
Q: How do you like to spend your free time?
I enjoy swimming whenever I can, especially a few times a week during the summer. Over the weekend, my wife and I like to bring our 10-month-old son to parks and museums (there are so many in Pittsburgh). I also love travelling to new places and trying all kinds of food along the way. Having spent years in Virginia and Michigan, I’m a fan of both the Virginia Cavaliers and Michigan Wolverines.