Statistical Science

A Conversaton with Leo Breiman

Richard Olshen

Full-text: Open access


Leo Breiman was born in New York City on January 27, 1928. His parents and he migrated five years later to San Francisco where he began school. During Leo's junior high school years, his family moved again, to Los Angeles. Leo graduated from Roosevelt High School in 1945 and entered the California Institute of Technology, from which he graduated four years later with a major in physics. He earned his Master's Degree in mathematics at Columbia University in 1950 and his Ph.D. in Mathematics at the University of California, Berkeley in 1954. Leo has broad ranging scientific and mathematical interests, including information theory and the theory of gambling. He has been involved in applications coming from studies of automobile traffic, air quality and toxic substance recognition. He is the author of a celebrated graduate text on probability theory, is one of four authors of Classification and Regression Trees and its associated CARTR software and has also written two other books. With Jerome Friedman, Leo developed the ACE (alternating conditional expectations) algorithm by which nonlinear relationships between the dependent variable and predictor variables in regression are described. He is the originator of “bagging” and “arcing,” both computer­intensive approaches to classification that are of much current interest.

Leo's professional positions have included being on the faculty of the Department of Mathematics at UCLA, an independent consultant for 13 years and Professor of Statistics and founding Director of the Statistical Computing Facility at the University of California, Berkeley. In addition, he has had visiting positions at Stanford and at Yale. For his many contributions, Leo has been honored by Fellowship in the Institute of Mathematical Statistics and in the American Statistical Association. He is an elected member of the American Academy of Arts and Sciences and received the Berkeley Citation from the University of California. The interests and accomplishments of Leo Breiman extend outside the areas of professional statistician and probabilist. He was a waiter in the Catskills, a dishwasher in the Merchant Marine, a trekker into the heart of rainforest Africa, an active father to many children from a small agrarian Mexican village, a member and President of the Santa Monica School Board, the architect of his stunning home and an accomplished sculptor. Leo and his wife, Mary Lou, reside in Berkeley. He is the father of two daughters, Rebecca and Jessica.

Article information

Statist. Sci., Volume 16, Issue 2 (2001), 184-198.

First available in Project Euclid: 24 December 2001

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier


Olshen, Richard. A Conversaton with Leo Breiman. Statist. Sci. 16 (2001), no. 2, 184--198. doi:10.1214/ss/1009213290.

Export citation


  • Breiman, L. (1957). The individual ergodic theorem of information theory. Ann. Math. Statist. 28 809-811. [Correction
  • (1960). Ann. Math. Statist. 31 809-810.]
  • Breiman, L. (1960). Optimal gambling systems for favorable games. Proc. Fourth Berkeley Symp. Math. Statist. Probab. 1 60-77. Univ. California Press.
  • Breiman, L. (1963). The Poisson tendency in traffic distribution. Ann. Math. Statist. 34 308-311.
  • Breiman, L. (1968). Probability Theory. Addison-Wesley, Reading, MA. [Republished (1991) in Classics of Mathematics. SIAM, Philadelphia.]
  • Breiman, L. (1991). The II-method for estimating multivariate functions from noisy data (with discussion). Technometrics. 33 125-160. (Awarded the Youden Prize as the best expository paper of the year in Technometrics.)
  • Breiman, L. (1992). Submodel selection and evaluation in regression-The X-fixed case and little bootstrap. J. Amer. Statist. Assoc. 87 734-751.
  • Breiman, L. (1994). The 1990 Census adjustment-Undercount or bad data? (with discussion). Statist. Sci. 9 458-475. Breiman, L. (1996a). Bagging predictors. Machine Learning 26 123-140
  • Breiman, L. (1998). Arcing classifiers (with discussion). Ann. Statist. 26 801-849.
  • Breiman, L. and Friedman, J.H. (1985). Estimating optimal transformations in multiple regression and correction (with discussion). J. Amer. Statist. Assoc. 80 580-619. (Theory and Methods Paper of the Year.) Breiman, L., Friedman, J. H., Olshen, R. A., and Stone,
  • C. J. (1984). Classification and Regression Trees. Wadsworth, Belmont, CA. (Since 1993 this book has been published by Chapman and Hall, New York.)
  • Freund, Y. and Schapire, R. (1996). Experiments with a new boosting algorithm. Machine Learning: Proceedings of the Thirteenth International Conference 148-156.
  • Freund, Y. and Schapire, R. (1997). A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. System Sci. 55 119-139.
  • Ji, C. and Ma, S. (1997). Combinations of weak classifiers. IEEE Trans. Neural Networks (Special Issue) 8 32-42.