Leo Breiman
leo@stat.Berkeley.EDULeo Breiman (Professor Emeritus) is is a member of the National Academy of Sciences. His current research interests are in computationally intensive multivariate analysis including the use of nonlinear methods for pattern recognition and prediction in high dimensional spaces. He is a co-author of Classification and Regression Trees and he has developed decision trees as computationally efficient alternatives to neural nets. This work has applications in speech and optical character recognition. He is the author of the textbooks Probability and Stochastic Processes with a View Toward Applications, Statistics with a View Toward Applications, and Probability.
Recent Papers
Some recent papers are available by FTP.
WALD Lectures
Three pdf files are available from the WALD lectures, presented at the 277th meeting of the Institute of Mathematical Statistics, held in Banff, Alberta, Canada (July 28 to July 31, 2002).
Machine Learning
Looking Inside the Black Box
Software for the Masses
Software Projects
Random Forests - updated March 3, 2004Survival Forests
