Olvi L. Mangasarian
John von Neumann Professor of Mathematics and Computer Sciences
Computer Sciences Department
University of Wisconsin
1210 W. Dayton St.
Madison, WI 53706-1685
telephone: (608) 262-1204
fax: (608) 262-9777
email:
olvi@cs.wisc.edu
http://www.cs.wisc.edu/~olvi/
Ph.D., Harvard University, 1959
Interests:
Mathematical programming, machine learning and data mining
Research Summary
Optimization theory is rich mathematically while being very effective
computationally in solving problems from diverse fields. My recent
work has emphasized mathematical programming approaches to machine
learning and data mining. I have used such approaches for training
neural networks, feature selection, parsimonious approximation,
clustering and massive data discrimination via support vector
machines. My mathematical programming interests include error
bounds for linear, nonlinear programs and variational inequalities,
ill-posed optimization problems, optimization problems with equilibrium
constraints as well as parallel distributed optimization algorithms
and smoothing techniques. Another important aspect of my research
is the use of mathematical programming techniques in diagnosing
breast cancer. This has resulted in an accurate computerized diagnostic
system which is in current use at University of Wisconsin Hospital.
Sample Recent Publications
Feature selection via concave minimization and support vector
machines (with P. Bradley), in Machine Learning: Proceedings
of the 15th International Conference (ICML '98), Jude Shavlik,
editor, pp. 82-90, Morgan Kaufmann, 1998.
Data mining: Overview and optimization opportunities (with
P. Bradley and U. Fayyad), to appear in INFORMS Journal on
Computing, 1999.
Regularized linear programs with equilibrium constraints,
in Reformulation-Nonsmooth, Piecewise Smooth, Semismooth and
Smoothing Methods, M. Fukushima and L. Qi, eds., pp. 259-268,
Kluwer Academic Publishers, 1998.
This page was automatically created December 30, 1998.
Email pubs@cs.wisc.edu
to report errors.