[BANANA] Seminar: Wed Dec 6

[Gene H Golub]

Linear Algebra and Optimization Seminar
Fall 2006
Gates Building, Room 104
Qing-Shan Jia
Wednesday, December 6, 4:15pm
Ordinal Optimization: Soft Optimization for Hard Problems

Ordinal optimization (OO) is an important tool to deal with 
simulation-based optimization problems, in which the performance of each 
design can only be evaluated through time-consuming simulation. Since 
first developed in 1992, ordinal optimization has become a research area 
with more than two hundreds published papers and many successful 
applications. This talk reviews the basic ideas of ordinal optimization 
and some recent major extensions, including:

    1. By comparing different selection rules and using a good one for the 
given problem, we are able to improve the efficiency of OO and further 
save the computing budget.
    2. By introducing the concept of layers in a multi-dimensional design 
space, vector ordinal optimization was developed to deal with multiple 
objective functions.
    3. By using a feasibility model, constrained ordinal optimization was 
developed to deal with constrained optimization.

As an example, we show the application of ordinal optimization in a 
remanufacturing system, in which OO saves the computing budget by at least 
one order of magnitude.

Biography
Qing-Shan Jia received the B.S. degree in automation and the Ph.D. degree 
in control science and engineering from Tsinghua University, Beijing, 
China, in 2002 and 2006 respectively. He is with Center for Intelligent 
and Networked Systems (CFINS), Department of Automation, Tsinghua 
University, and is also visiting the Division of Engineering and Applied 
Sciences, Harvard University. His current research interests are 
simulation-based optimization and agent-based decentralized optimization 
of complex systems, with a specific focus on ordinal optimization.


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++



Gene Golub,  Fletcher Jones Professor of Computer Science
Gates 2B
Computer Science Dept
Stanford University
Stanford, CA 94305
USA

Office Phone: 650 723 3124
Home Phone: 650 323 0105
FAX:  (650) 322 3277
Mobile: 650 796 5402