Loading timeline…
20012000s
Jerome H. Friedman
Stanford statistician who turned boosting into gradient descent in function space, producing the method that still wins on tabular data.
Organizations
Stanford UniversitySLAC National Accelerator Laboratory
Major Achievements
- •Formulated gradient boosting (2001), recasting boosting as optimisation in function space.
- •Co-created CART, the classification and regression tree framework, with Leo Breiman.
- •Co-wrote 'The Elements of Statistical Learning', a standard text of the field.
- •Developed projection pursuit and multivariate adaptive regression splines.
Key Papers
- Greedy Function Approximation: A Gradient Boosting Machine
Jerome H. Friedman · The Annals of Statistics · 2001