David Allen McAllester
David Allen McAllester’s work connects formal reasoning with machine learning. His work includes planning research, PAC-Bayesian theory, and shared object-detection systems. McAllester completed his MIT doctorate in 1987, served at Cornell in 1987–1988 and worked at MIT from 1988 to 1995. His 1998 work developed PAC-Bayesian theorems. These connect Bayesian and non-Bayesian methods in the study of machine learning.
- Works
- 183
- Citations
- 29,024
- h-index
- 56
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1987–1995
Academic research
McAllester completed his MIT doctorate in 1987, served at Cornell in 1987–1988 and worked at MIT from 1988 to 1995.
-
1997
Research recognition
AAAI elected David Allen McAllester as a Fellow in 1997.
-
1998
Learning theory
His 1998 work developed PAC-Bayesian theorems. These connect Bayesian and non-Bayesian methods in the study of machine learning.
-
2002–2017
Building a research institute
He served as TTIC’s chief academic officer from 2002 to 2017.
-
2007–2009
Shared object-detection research
McAllester, Pedro Felzenszwalb, Ross Girshick, and Deva Ramanan developed a deformable-part object-detection system. The system achieved strong PASCAL challenge results from 2007 to 2009.
Sources and credits
Photo: Artist: Photo by JasonSmith.com; Copyright: http://www.jasonsmith.com; supplied by www.ttic.edu. License: Not stated by source. Source: ttic.edu. Cropped to a square and resized. Ask us to remove or replace this photo.