Can you tell us a bit about your background and how you decided on statistics?
I’m originally from southeastern China, and I’m the only child in my family. Growing up, my mom and my grandma encouraged me to be independent and to seek a career I would enjoy and fight for my entire life. When I took the exam to go to college, I got full marks in mathematics. This gave me the confidence to choose a mathematics-related major as an undergraduate student at Zhejiang University.
What inspired me to do a deep dive into the statistics major was the Fourth Industrial Revolution happening during that time. Related to the realization of Society version 5.0, we’re in the process of imagining some new society where we will heavily rely on big data to solve challenges and to figure out how people can use artificial intelligence to live better. Therefore, I decided to get a minor in the Advanced Honor Class of Engineering Education at Chu Kochen Honors College, which granted me great opportunities to work with outstanding peers across the university and learn cutting-edge technologies to integrate statistics and machine learning elegantly.
What motivated you to then earn your Ph.D.?
During my undergraduate, I attended several mathematical and machine learning competitions and won the Meritorious Winner in American Mathematical Contest. Yet I could still see my limitations in knowledge, which motivated me to reach for the highest standard and become an expert in statistics and machine learning. After a summer research internship at the University of Alberta in Canada, I decided to go abroad for higher education and came to the United States after earning my bachelor’s degree.
I spent five years at North Carolina State University and got my Ph.D. in statistics under the co-supervision of Dr. Wenbin Lu and Dr. Rui Song. My dissertation was on optimal decision-making and policy evaluation with complex data, which leverage causal inference and reinforcement learning to help us make the right decisions for the right people at the right time. During that time, I also spent three years interning at Amazon. That was another excellent opportunity for me to see the real-world impact of computer science and statistics and how our research contributes to industry and society.
Can you talk about your research focus?
I have broad research interests in methodology and theory in causal inference, reinforcement learning and graphical modeling, and in how they interact to establish reliable, powerful and interpretable solutions to a variety of real-world problems. Currently, I’m working in two main directions. One is personalized decision-making and policy evaluation for complex data. This area integrates the techniques of causal inference and reinforcement learning, which enables us to answer counterfactual questions so we can make better decisions in an interactive environment. Some use cases include precision medicine and personalized recommendations in e-commerce. I’m also very interested in causal discovery with emerging challenges in big data. This area leverages causal inference and graphical models to disentangle the complex relationship among variables and deliver the results in an explainable way. My long-term research goal is to teach a machine to learn the logic from the facts.
So, what brought you to UCI?
First of all, I love Irvine. This city is vivid, elegant and multicultural, with a lot of young tech companies, presenting appealing opportunities for collaboration. The second reason is owing to the great research communities inside UCI and around the UC system. There are so many outstanding faculty members here I would like to work with, in my department, in the School of ICS, and across the university. Also, where else can we find the Departments of Statistics, Computer Science, and Informatics, all in the same school? It’s a very unique place.
What courses will you be teaching?
I will be teaching a course in spring 2023 for undergraduates called “Introduction to Probability and Statistics.” In the future, I would like to design some new courses on reinforcement/machine learning and on causal discovery.
Is there a book you’d like to recommend to students?
The Book of Why by Judea Pearl and Dana Mackenzie. This book is for non-technical people who are interested in causal discovery, explaining how we understand or formulate causality in real problems, and highlighting the difference between correlation and causation. It helps us understand the logic of things happening, dive into the essence of human thought, and use the causal principle as the key to applying artificial intelligence to solve real problems in meaningful ways.
What do you do in your spare time?
I like hiking, going to the beach and being out in the sunshine. I also like to go kayaking and standup paddle boarding. If I have a bit more time, I like to travel to new places. I’ve only been in Irvine for a few months, and I’ve already been to Las Vegas, Death Valley, Antelope Valley and San Francisco. Then from San Francisco, I drove down the coast to Los Angeles. Next, I plan to visit San Diego and Yosemite. Work hard, play hard! I enjoy different views, which open my eyes and my mind. I also like to watch movies to experience different lives. One of my favorites is “Forrest Gump.”
Finally, what do you like best so far about UCI?
The people here are so nice! I have received a lot of support from my department chair, faculty members and department staff during my relocation and onboarding, which is extremely helpful for the early stage of my career. There are also plenty of opportunities for collaborating, and I have been invited to give some talks. For example, I gave a talk in early September on “Optimal Subgroup Identification via Constrained Policy Tree Search” for CHOC Research. I will continue to explore the fantastic research community at and around UCI as I study how statistics and machine learning can uniquely contribute to interdisciplinary solutions. And I will continue working with all these brilliant people to help build a brighter future.
— Shani Murray