Department of Statistics Unitmark
Dietrich College of Humanities and Social Sciences

In honor of the Department of Statistics' 50th Anniversary, we are highlighting our outstanding faculty and alumni of the past half-century.

Larry Wasserman

Professor Larry Wasserman, who was recently elected to the National Academy of Sciences (NAS), enrolled at the University of Toronto as an undergraduate to study computer science. “They had punch cards then and I couldn’t stand them, so I switched to statistics,” he said. As he was about to complete his Ph.D. (for which he later won the Pierre Robillard Award for the best thesis in probability or statistics), Carnegie Mellon came calling.  “I was sitting in my apartment in Toronto in 1987 working on my thesis when the phone rang.   ‘This is Steve Fienberg.  Mind if I come over for a cup of coffee?’  “A phone call from a famous statistician was the last thing I expected.  What could I say but, ‘Sure.  Come on over.’ “Steve immediately put me at ease.  We had a nice chat about CMU, Pittsburgh, and probably a few other things that I now forget.  I had a strong sense that the CMU Statistics Dept. would be a nice place to visit,” he said.

 Once here, his first impressions were reflective of a new kid on the block.  “Everyone was so old in attitude, but friendly.  “I felt like an eighteen-year-old, and they were all grown-ups to me,” he said. The impression from the senior faculty was of a potential star.  “Larry came here on a Canadian NSERC Fellowship.  The plan was: nine months here and nine months at Stanford,” Professor Jay Kadane recalled.  “After six months, he decided he was having such a good time he was not going to Stanford. “After twelve months, we decided he was so good we were not going to let him go without a fight. "And he is still here,” Jay said.
 
Larry’s research spans both theoretical and applied statistics, with the theoretical side focusing on the intersection of statistics and machine learning.  His work has provided new methods and theory for simultaneously estimating the relationships between large numbers of variables and for finding subtle spatial structure in complex datasets. He has made fundamental contributions to the theory of statistical inference; developed clever and effective methods for data analysis; and advanced the state of the art in machine learning.

 In the late 1990s Larry was one of the first faculty to engage in a research center that evolved into CMU’s Machine Learning Department. As Prof. Feinberg wrote in a recent issue of the IMS (Institute of Mathematical Statistics) Bulletin: “Larry’s early research led to his recognition as a leading Bayesian innovator who has integrated insights into the foundations of inference with incisive and applicable methodological contributions. “His contributions range from definitive treatments of Bayesian robustness and modern nonparametric estimation, mixture models, multiple testing, privacy, and causal inference, and highly successful collaborations with astrophysicists and statistical geneticists. “His legions of collaborators come from across his departments at CMU and across multiple fields and countries.”
 
In 1996, Larry was named a Fellow of the American Statistical Association and of the IMS. In 1999, he received the Presidents’ Award of the Committee of Presidents of Statistical Societies for the Outstanding Statistician Under the Age of 40.  In 2002, he was awarded the Centre de Recherches Mathematique de Montreal-Statistical Society of Canada Prize in Statistics for his original contributions to statistical theory and his development of Bayesian methodology.  Larry is the author of two highly praised textbooks: All of Statistics (winner of the 2005 DeGroot Prize from the International Society for Bayesian Analysis), and All of Nonparametric Statistics. In 2013, he presented the prestigious IMS Rietz Lecture on topological inference.
 
Closer to home, Larry is a stellar workplace citizen and colleague. “Beyond his spectacular research accomplishments, Larry has been a pillar of our department,” Professor Robert Kass said. “I would point to his inspiring lectures introducing theoretical statistics to graduate students, which led to his book, All of Statistics, and to his demonstration of the continued role of statistics in machine learning, partly through teaching his core course for our Machine Learning Department. From the beginning, Larry has poured his efforts into the most interesting problems in statistical theory, without losing sight of what matters to statistical practice. “For several years, while Larry was an assistant professor, he was also my next-door neighbor in our department. We talked at length nearly every day, and soon he had soaked up all the knowledge I had to offer. I take great pride in having co-authored with him several of his (and my) most highly cited papers, and in whatever role I may have played as he began his development toward becoming a preeminent statistician. Our department provided, and continues to provide, a wonderfully collaborative environment, and Larry has become a stellar role model for all to follow,” he said.

 “Larry embodies the ideal Carnegie Mellon Dept. of Statistics faculty member,” Professor John Lehoczky said.  “He is a leader in core fields of statistics; his innovative work pushes the boundaries of the field; he is major contributor to scientific problems outside of statistics (e.g. astronomy and cosmology); and he brings the new methodologies that he develops in those fields back to be incorporated into statistical methodology. “Such accomplishments are vital for the health of the field of statistics, and very rare,” he said. “Larry has made incredible contributions to a wide range of problems in statistics,” Professor Sivaraman Balakrishnan said. “Just in the last five years he has made enormously influential contributions to differential privacy, high-dimensional statistics, non-parametric statistics, conformal prediction, statistical inference, topological data analysis and machine learning. “He is one of the great unifiers of the fields of statistics and machine learning, effortlessly crossing disciplinary boundaries and always keen to learn about new models, new ideas and new perspectives. To me personally, he was my first statistics teacher, and has tirelessly continued educating me since. His openness, energy and infectious enthusiasm have inspired me, and several other young researchers, to develop a deep appreciation for the beauty of statistics,” he said.

"Larry is all this, and he also has a wicked sense of humour, strong opinions about politics, economics, string theory, etc.,” Professor Valérie Ventura said “It's always fun to hang out and argue with him. Probably the only thing he is really bad at is reducing his carbon footprint,” she said. “Probably I am the person who knows Larry best,” Professor Isa Verdinelli said. “I know how passionate he is about statistics, and the amount of fun he gets from his work.  He is also on the funniest people one might meet, always read to crack a joke or do hilarious impersonations. “And then there is Larry the flute player, Larry the painter, Larry the couch potato, Larry the TV shows fan, and Larry the proud father of cats. “When I met Larry he was a young kid, just moved to CMU.  He I still a kid, ready to play tricks, and to laugh with all of us,” she said.

The NAS induction ceremony will be April 29, 2017, at which time Larry will join 84 new members and 21 foreign associates from 14 countries elected into the NAS this year. “It’s a great honor, and I get to go to D.C. and go to a ball. “The bad news is I have to get a tux for the ball,” he said. As for the future, Larry plans “to keep doing what I do. “I don’t know anything else,” he said.