Peter van Beek
Peter van Beek earned his PhD at Waterloo in 1990 and then joined the University of Alberta. He returned to Waterloo ten years later. His research spans knowledge representation, constraint satisfaction and applied machine learning. In constraint programming, acceptable solutions are described with constraints and a search procedure finds a solution that satisfies them. He has applied this view to learning Bayesian network structures and decision trees. His projects also cover water-use analysis and computational photography. He served as editor of Constraints from 2005 to 2009.
- Works
- 135
- Citations
- 4,577
- h-index
- 35
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
1984–1990
Training and an Alberta appointment
He earned a 1984 BSc at UBC, a 1986 M.Math at Waterloo and a 1990 PhD at Waterloo. He joined Alberta’s faculty after the PhD and returned to Waterloo ten years later.
-
2005–2009
A forum for constraint research
Van Beek served as Editor-in-Chief of Constraints from 2005 to 2009. Van Beek describes constraint programming as specifying acceptable solutions and using search to satisfy those constraints, across a range of optimization and reasoning problems.
-
As of October 2026
Search for machine learning models
He studies Bayesian network structure learning and decision tree learning as difficult combinatorial optimization tasks. His group applies constraint programming and advanced search techniques.
-
As of October 2026
Water data and computational photography
His projects include analysis of hourly smart water meter records for leak detection and forecasting. He also studies supervised learning methods for autofocus and focus stacking, where images at different focus depths are combined into a single image.
Sources and credits
Photo: uwaterloo.ca; creator not stated. License: Not stated by source. Source: uwaterloo.ca. Cropped to a square and resized. Ask us to remove or replace this photo.