|
Academic
Biography
Shai Ben-David grew up in Jerusalem, Israel. He attended the Hebrew University
studying physics, mathematics and psychology. He received his PhD under the
supervision of Saharon Shelah and Menachem Magidor for a thesis in set theory
(on non-provability of infinite combinatorial statements).
Dr. Ben-David was a postdoctoral fellow at the University of Toronto
in the Mathematics and the Computer Science departments, and in 1987 joined
the faculty of the CS Department at the Technion (Israel Institute
Technology). He held visiting faculty positions at the Australian
National University
in Canberra (1997-8) and at Cornell University
(2001-2004). In August 2004 he joined the School
of Computer Science at the University of Waterloo.
Research Interests
My 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 interplay between mathematical theories and
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 clear 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 Learning-To-Learn paradigm (providing
some theoretical justifications to common practices),
at the Stability method for determining the number of clusters in a data set
(see my COLT06, COLT07 papers on that and a recent submission),
and quite a few more topics.
Clustering is a really
large area that also suffers from the lack of mathematical foundations, and I
have been working extensively trying to address the theoretical challenges of developing such a theory (see my
recent submissions with Ackerman as well as my ’05 paper with von
Luxburg)
For more details see the links to my papers, talks, and professional
activities.
2009 Publications
Learning Low-Density Separators
S. Ben-David, Tyler Lu, David Pal, and
Miroslava Stakova. Proceedings of
the Twelfth International Conference on
Artificial Intelligence and Statistics, 2009.
A Uniqueness Theorem for Clustering.
Reza Bosagh Zadeh, Shai Ben-David. Proceedings
of UAI 2009.
Agnostic Online Learning.
Shai Ben-David, David Pal and Shai
Shalev-Shwartz. Proceedings of COLT 2009.
Modeling and Querying Possible Repairs in
Duplicate Detection
George Beskales, Mohamed A. Soliman, Ihab F. Ilyas and Shai Ben-David
To Appear in Proceedings of VLDB 2009.
2008 Publications
Measures of Clustering Quality: A Working
Set of Axioms for Clustering
Margareta Ackerman and Shai Ben-David, Proceedings of Neural
Information Processing Systems (NIPS 2008)
Does Unlabeled Data Provably Help? Worst-case Analysis of the
Sample Complexity of Semi-Supervised Learning
S.Ben-David, Tyler Lu and David Pal. To appear in COLT 2008
Relating clustering stability to properties of cluster boundaries
S. Ben-David and Ulrike von Luxburg. To appear in COLT 2008
A Notion of Task relatedness Yiealding Provable
Multiple-task Learning Guarantees
S. Ben-David and Reba Schuller-Borbely. (Machine Learning Journal,
2008).
A framework for addressing the Training/Test
Distributions Gap
S. Ben-David. Book chapter. To appear in ``Data Shift in Machine
Learning'' MIT press.
2006-7
Publications
A Framework for
Statistical Clustering with Constant Time Approximation for K-Means
Clustering
S. Ben-David. (Machine Learning Journal, 2007)
Stability
of K-Means
S. Ben-David, D. Pal, H. ulrich Simon(Proceedings of COLT07)
Analysis
of Representations for Domain Adaptation
S. Ben-David, J. Blitzer, K. Crammer, F. Pereira (Proceedings of
NIPS'06)
Learning
Bounds for Support Vector Machines with Learned Kernels
Nathan Srebro, Shai Ben-David (Proceedings of COLT'06)
A Sober Look at
Stability of Clustering
S. Ben-David, U. von Luxburg, D. Pal Winner
of Best Student Paper Award in COLT'06
Alternative
Measures of Computational Complexity
S. Ben-David TAMC'06
Towards a Statistical Theory of Clustering
(with Ulrike von Luxburg) - PASCAL
Workshop on Statistics and Optimization of Clustering (2005)
Non-Parametric Change Detection in 2D Random
Sensor Fields
(with Ting He and Lang Tong) Winner of Best
Student Paper Award in ICASSP 2005
|
