Leo Breiman

Leo Breiman ( born January 27, 1928 in New York City; † July 5, 2005 in Berkeley, California ) was an American statistician who worked at the University of California, Berkeley.

Breiman was the only son of Eastern European immigrants Max and Lena Breiman, his father was a tailor. He grew up in San Francisco and Los Angeles, where he attended Roosevelt High School until 1945. In 1949 he successfully finished studying physics at Caltech, in 1950, he earned a master's degree in mathematics at Columbia University, 1954, he graduated with a Ph.D. then at the University of California at Berkeley and taught there from probability theory. Appeared to him the education of his students to practice foreign, he taught mathematics at times even in a school. After a few years he finished his teaching career, first authored a textbook on probability theory and then worked as a consultant for statistics, also for UNESCO in Liberia. From 1980, he again worked at UCLA, where he established a department for computer-assisted statistics calculation, which could take on tasks for other institutions later. In 1993 he finished his work, but still supervised some students and published some of his best known works only after the end of his official teaching (including Random Forest ).

Breiman worked at the interface between computer science and statistics, particularly in the field of machine learning. His most significant contributions are the work on decision trees (CART algorithm for decision-making: Classification and Regression Trees ) and compilations of decision trees to bootstrap samples. He called this bootstrap compilations Bagging. Another of him with developed method called Random Forest.

Breiman has received many awards and received prizes. He was elected to the United States and the American Academy of Arts and Sciences as a member of the National Academy of Sciences.