Samuel Kou Wins 2012 COPSS Presidents’ Award
The Committee of Presidents of Statistical Societies recently recognized Samuel Kou for groundbreaking contributions to stochastic modeling and statistical inference in single molecule biophysics; for pioneering the equi-energy sampler; for fundamental contributions to Bayesian, empirical Bayes, and nonparametric methods; and for outstanding service to the statistical profession and contribution to statistical education.
Kou was born in Lanzhou, China, and attended Lanzhou No. 1 High School. He earned his bachelor’s degree in computational mathematics from Peking University in 1997 and his PhD in statistics from Stanford University in 2001 under the supervision of Bradley Efron. He then joined Harvard University as an assistant professor and was promoted to full professor in 2008.
He is the recipient of the National Science Foundation’s CAREER Award, the Institute of Mathematical Statistics’ Richard Tweedie Award, and the Raymond J. Carroll Young Investigator Award. He is an elected Fellow of the American Statistical Association and Institute of Mathematical Statistics (IMS) and an elected member of the International Statistical Institute. He was an IMS Medallion Lecturer in 2009 and the recipient of the 2010 ASA Outstanding Statistical Application Award. He is currently an editor of the Annals of Applied Statistics and CHANCE magazine, in addition to serving/having served as an associate editor for Statistical Science, Journal of the American Statistical Association (Applications and Case Studies), Bernoulli Journal, Annals of Applied Statistics, Journal of Multivariate Analysis, and Statistica Sinica.
Kou has made fundamental contributions to interdisciplinary, methodological, computational, and mathematical statistics through his work on Bayesian and Monte Carlo methods, nonparametric methods, stochastic modeling and inference in biophysics, and stochastic modeling in finance and economics. He actively works in the interdisciplinary field of nanoscale biophysics, a booming field that lies at the intersection of biology, chemistry, physics, and nanotechnology. His groundbreaking contributions include the first likelihood-based Bayesian data augmentation method for handling unobserved molecular Brownian diffusion in single-molecule experiments; the first fractional Brownian motion–based Hamiltonian model to explain the experimentally observed conformational fluctuation of protein molecules, which defied the classical Einstein Brownian diffusion model; and the first microscopic stochastic model that successfully explains the previously mysterious experimental findings in enzymatic reactions.
Kou’s seminal contributions to Monte Carlo methods include the equi-energy sampler, a new Monte Carlo framework for statistical sampling and inference that has significantly improved the efficiency of Monte Carlo simulation and provided a fundamentally new perspective on study sampling and inference problems; the multiresolution method—a general inference framework—to study diffusion processes; and the sequential Monte Carlo method FRESS (fragment regrowth via energy-guided sequential sampling) to study protein folding.
These methods have been successfully applied to a range of challenging scientific problems, such as DNA motif finding in computational biology, thermodynamic estimation in statistical mechanics, exploring the energy landscape of biomolecules, and protein folding.
Kou’s work in nonparametric and empirical Bayes methods include using curved exponential family to construct model-selection criterion, SURE (Stein’s unbiased risk estimate) inference of hierarchical models, and kernel estimation of doubly stochastic Poisson processes.
Kou is a well-respected teacher who has played an important role in training students at Harvard by first serving as head tutor and now co-director of graduate studies.
Following is an interview between Kou and Bhramar Mukherjee, COPSS secretary/treasurer:
What was your first reaction to winning the prestigious COPSS President’s Award?
I remember I was in my office preparing the lecture notes for the next classes, when Tony Cai, the committee chair, called, informing me about the news. I was quite surprised, because it was still in February and I thought the result would not be known until much later. And, of course, I was overjoyed—it is such a great honor. I then shared the news with my parents and brother. They were all very happy.
Which parts of your job do you like the most?
I enjoy being a statistician, working on a diverse range of problems. Many of the problems I work on come from real scientific issues: the challenge to analyze experimental data or the challenge to construct a stochastic model to explain the experimental puzzles. Along the way, I get the chance to learn some science. As John Tukey said, “The best thing about being a statistician is that you get to play in everyone’s backyard.”
What advice would you give to young people who are entering the profession as PhD students and assistant professors at this time?
Following the heart, not the trend. I think only by doing what you love to do, by listening to your heart, can you proceed forward and enjoy the process and the fruits in the long run. I feel it is worth keeping in mind that what is hot right now might not be so hot five or 10 years down the line.
Who are your most significant mentors? How did/do they affect your career?
I was fortunate to be a student of professor Bradley Efron. I learned from him not only statistics, but more importantly, how to approach a problem and how to think. Later on, at Harvard, I was lucky to get the chance to learn from and be mentored by professors Don Rubin, Wing Wong, and Jun Liu. Professor Sunney Xie in the department of chemistry and chemical biology of Harvard University is my science mentor. I picked up from him not only some chemistry, biology, and physics, but also a perspective of science and the role of statistics in science.
How did you start to work on statistical problems in biophysics?
Not long after I joined the Harvard Statistics Department, I met professor Sunney Xie at a lunch (arranged by professor Jun Liu). He told me that his group was doing experiments on single molecules to study biological systems. I was immediately fascinated by it, though I had never heard of single-molecule experiments before. I asked him for some background papers to read. It took me a few months to get a rudimentary understanding of what they were doing. I realized that there were a lot of exciting statistical problems in the field because, fundamentally, at the single-molecule level, everything is stochastic, where statistics has an important role to play. I then started to collaborate with professor Sunney Xie.
Anything else you will like to share about our profession?
I love statistics. A statistician is like a 19th-century mathematician: On one hand, they were working on problems in mechanics, fluid dynamics, optics, astronomy, etc., and on the other hand, they were working on theories, structures, and methods. I guess, in the modern world, statistics is one of the very few disciplines that still enjoy such interplay.
Finally, what are your hobbies/interests beyond statistics?
I like to travel and explore different cultures. Working in academia is wonderful in this regard. I like outdoor hiking, though I am not a serious hiker. I like reading books on history, philosophy, and general-audience science.