Jerome H. Friedman

Timeline

1. 1939–1968

   Physics at Berkeley

   Born in Yreka, California, in 1939, he spends two years at Chico State College and transfers to the University of California, Berkeley in 1959. He earns an AB in physics in 1962 and a PhD in high-energy particle physics, then starts work as a research physicist at the Lawrence Berkeley National Laboratory in 1968.

2. 1972–1991

   From SLAC to Stanford statistics

   In 1972, he becomes leader of the Computation Research Group at the Stanford Linear Accelerator Center, and in 1974 he publishes a projection pursuit algorithm for exploratory data analysis with John Tukey. Stanford appoints him Professor of Statistics in 1982, and the American Statistical Association elects him a Fellow in 1984. He chairs Stanford's Department of Statistics from 1988 to 1991.

3. 1975–1984

   Decision trees with CART

   He first meets Leo Breiman in 1975, when both are independently investigating CART principles. In 1977, he coauthors several algorithms for speeding up nearest-neighbor classifiers. With Breiman, Richard Olshen and Charles Stone, he produces the 1984 book Classification and Regression Trees and releases the commercial product CART, which introduces the gini and twoing splitting rules, cost-complexity pruning and cross validation for decision trees.

4. 1991–1997

   MARS and PRIM

   In 1991, he extends recursive partitioning ideas to regression with Multivariate Adaptive Regression Splines (MARS). He introduces the Patient Rule Induction Method (PRIM) in 1997, an interactive data mining method he describes as "Bump Hunting in High Dimensional Data."

5. 1999–2001

   Gradient boosting and a textbook

   With Trevor Hastie and Rob Tibshirani, he shows that boosting is a form of additive logistic regression, and he develops explicit regression gradient boosting algorithms in 1999 and 2001. His paper "Greedy function approximation: a gradient boosting machine" appears in the Annals of Statistics in 2001. In 2001, he also co-writes The Elements of Statistical Learning: Data Mining, Inference, and Prediction with Hastie and Tibshirani, a book on the importance of statistics in data and machine learning.

6. 2002–2017

   SIGKDD award and NAS election

   He wins the ACM SIGKDD Innovation Award in 2002 and the Emanuel and Carol Parzen Award for Statistical Innovation in 2004, and becomes a fellow of the American Academy of Arts and Sciences in 2005. The National Academy of Sciences elects him a member in April 2010, and he receives the 2012 IEEE ICDM Research Contributions Award. In 2017, Stanford's Department of Statistics renames its applied statistics thesis prize for him.

7. Since 2007

   Professor emeritus at Stanford

   He becomes Professor Emeritus in Stanford's Department of Statistics in 2007, and his primary research interest is machine learning. In May 2019, the department marks his 80th birthday with a day-long symposium at the Huang Engineering Center.

8. 2023–2025

   Stanford's consulting workshop

   As Professor of Statistics, Emeritus, he teaches Stanford's Consulting Workshop, STATS 390, in spring of the 2023–24 and 2024–25 academic years.