Persi Diaconis

Mary Sunseri professor of Mathematics and Statistics, Stanford University

Persi Diaconis is universally acclaimed as one of the world's most distinguished scholars in the fields of statistics and probability. His work ranges widely from the most applied statistics to the most abstract probability. His outstanding intellectual versatility is combined with an extraordinary ability to communicate in an entertaining and accessible manner. His talks are engaging, beautiful, and diverse, and he is therefore constantly in demand as a speaker.

Born into a family of professional musicians, Diaconis studied violin at Julliard from the ages of 5 to 14. At 5 years of age the young Persi also picked up a book on magic and learned tricks form it on his own. When the greatest magician in the US invited Persi to join him on the road, Persi could not resist, and left home at age 14 without telling his parents. His 10 years as a professional magician led Diaconis to study probability, and within five years he had entered college and earned a PhD in mathematics (Harvard, 1974). His work of the next few years quickly attracted international attention. It already spanned statistics, probability and mathematics with applications to medicine, ESP research, computer science, vision, cognitive illusions, and more. In 1982 Diaconis was among the earliest winners of a MacArthur genius Fellowship. He has held tenured positions at Harvard, Cornell, and Stanford, where he currently teaches. He is a Fellow of the American Academy or Arts and Sciences and a Member of the National Academy of Sciences.

Diaconis's research is as deep as it is diverse. Among the highlights is his pioneering work on the speed of convergence of Markov chains to equilibrium, a rapidly growing field with numerous applications to statistics, physics, computer science, and other areas. His dramatic and famous cut-off phenomenon has been nothing short of amazing. Together with David Freedman of Berkeley, Diaconis has made fundamental, precise, and dramatic contributions to Bayesian statistics.

But the impact of his contributions extends far beyond probability and statistics: beyond the academic applications of his work, Diaconis's unique background as both a magician and a statistician has enabled him to debunk with unusual authority much research on ESP and the so-called paranormal. He has exposed several psychics, including Uri Geller. Recently he wrote the entry on magic for the authoritative Routledge Encyclopedia of Philosophy.

Diaconis's first Patten lecture, "On Coincidences of Everyday Life", has been a topic of his attention for decades. Coincidences can affect where we live and what we do. After reviewing the early work of Jung and Freud, Diaconis will show with a bit of quantitative thinking that coincidences are not so surprising after all. By contrast, "The Search for Randomness" will take a careful look at some of our most primitive examples of random phenomena: tossing a coin, throwing a dart at the wall, and shuffling cards. In each case, while randomness can be achieved, normally there are substantial biases. A careful look at the foundations can help protect us from error in census adjustment, random number generation and mindless statistical modeling.