| Manfred Klaus Warmuth | |
| Field: | Computer Science |
| Work Institution: | |
| Alma Mater: | University of Colorado, Boulder |
| Doctoral Advisor: | Hal Gabow |
| Doctoral Students: | Yoav Freund |
| Known For: |
|
| Prizes: | Elected to Leopoldina (2021) |
Manfred Klaus Warmuth is a computer scientist known for his pioneering research in computational learning theory. He is a Distinguished Professor emeritus at the University of California, Santa Cruz.
After studying computer science at the University of Erlangen–Nuremberg, earning a diploma in 1978, Warmuth went to the University of Colorado Boulder for graduate study, earning a master's degree there in 1980 and completing his Ph.D. in 1981. His doctoral dissertation, Scheduling on Profiles of Constant Breadth, was supervised by Harold N. Gabow.
After postdoctoral research at the University of California, Berkeley and Hebrew University of Jerusalem, Warmuth joined the University of California, Santa Cruz in 1983, became Distinguished Professor there in 2017, and retired as a professor emeritus in 2018. He was a visiting faculty member at Google Brain from 2019 to 2020.
With his student Nick Littlestone, Warmuth published the weighted majority algorithm for combining the results for multiple predictors in 1989.
Warmuth was also the coauthor of an influential 1989 paper in the Journal of the ACM, with Anselm Blumer, Andrzej Ehrenfeucht, David Haussler, introducing the Vapnik–Chervonenkis dimension to computational learning theory. With the same authors, he also introduced Occam learning in 1987.
In 2021, Warmuth became a member of the German National Academy of Sciences Leopoldina.