Professor Ben-David's research interests span a wide spectrum of topics in the foundations of computer science and its applications, with a particular emphasis on statistical and computational machine learning. The common thread throughout my research is the quest for mathematical foundations of real world problems. In recent years much of my research has been directed towards providing mathematical analysis for popular machine learning and data mining paradigms that seem to lack solid theoretical justification. I have looked into the performance guarantees one can provide for Support Vector Machines (with pessimistic conclusions), at Semi-Supervised Learning (once again, coming up with some inherent limitations of that approach), at the problem of domain adaptation paradigm (providing the first theoretical justifications to common practices), change detection in streaming data, at the Stability method for determining the number of clusters in a data set (see my COLT06, COLT07 and COLT08 papers on that issue), and quite a few more topics. Clustering is a very wide research area, with many practical application, that also suffers from the lack of mathematical foundations. I have been working extensively trying to address the challenge of developing a theory that provides guidelines for choosing an appropriate clustering technique for a given task (see my recent papers with Ackerman et al).
Degrees and awards
BSc, MSc, (psychology), BSc, MSc, PhD (math) (Hebrew University)
President of the Association for Computational Learning (2009-2011); Ross and Muriel Cheriton Faculty Fellowship, University of Waterloo (2008-2011)
Industrial and sabbatical experience
Professor Ben-David has been a postdoctoral fellow at the University of Toronto, in both the Mathematics and Computer Science Departments. He then joined the Computer Science department at the Technion, Haifa, in Israel where he served as an associate professor until 2004. In 1996-7 Professor Ben-David spent a sabbatical year in the Australian Institute for Advanced Studies at the Australian National University in Canberra, and in 2001-4 he spent three years as a visiting professor at Cornell in the Computer Science department and in the school of Electrical and Computer Engineering.
A Theory of Learning from Different Domains. Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jenn Wortman. Machine Learning 79(1-2):151-175 (2010)
Shai Ben-David, Margareta Ackerman: Measures of Clustering Quality: A Working Set of Axioms for Clustering. NIPS 2008:121-128
S. Ben-David, U. von Luxburg, and D. Pal. A Sober Look at Clustering Stability. Winner of Best Student Paper Award. Proceedings of Conference on Learning Theory (COLT), 2006.
S. Ben-David. Alternative Measures of Computational Complexity. Proceedings of Conference on Theory and Applications of Models of Computation (TAMC), 2006.
T. He, S. Ben-David, and L. Tong. Non-Parametric Change Detection in 2D Random Sensor Fields. Winner of Best Student Paper Award. Proceedings of ICASSP, 2005.
D. Kifer, S. Ben-David, and J. Gehrke. Identifying Distribution Change in Data Streams. Proceedings of Very Large Data Bases, pp. 180-191, 2004.
S. Ben-David, N. Eiron, and H. Simon. Limitations of Learning via Embeddings in Euclidean Half-Spaces Journal of Machine Learning Research, 3:441-461, 2002.