RESEARCH INTERESTS
My
research
interests are mainly in continuous optimization. More
specifically, I am interested in interior-point algorithms for convex
optimization.
My thesis advisor was Michael
J.
Todd.
During
my doctoral study at the Operations
Research
department at Cornell
University,
I have developed a new approach to sensitivity analysis in linear and
semidefinite programming that relies entirely on interior-point
algorithms.
This approach eliminates the need to cross over to a basic optimal
solution
in linear programming and circumvents the drawbacks of the classical
optimal
basis approach. Furthermore, this is one of the first practical
approaches
to sensitivity analysis in semidefinite programming. I have also worked
on warm-start strategies in interior-point methods for linear
programming
with Stephen J. Wright.
I proposed a unifying perspective on the optimal partition approach to
sensitivity analysis in conic optimization. Currently, I am interested
in the design and analysis of approximation algorithms for certain
large-scale
convex optimization problems that have a special structure such as
geometric optimization problems.
RESEARCH GRANTS
- Role: Principal Investigator, Type of Grant: National Science Foundation Faculty Early Career Development (CAREER)
Award, Funding Agency: National Science Foundation, Title: CAREER: A Unifying Interior-Point Approach to Sensitivity Analysis and Reoptimization in Conic Programming, Amount: US$400,000, Duration: June 1, 2003 -- May 31, 2008 (terminated in August 2006 due to relocation), Project Number: DMI0237415. (Abstract)
- Role: Principal Investigator, Type of Grant: Strategic Partnership for Industrial Resurgence (SPIR), Funding Agency: ADP Investor Communications Services, Title: SPIR - Algorithmic Aspects of Booklet Bundling, Amount: US$18,593, Duration: July 15, 2004 -- July 15, 2005. (Co-PIs: Joseph S. B. Mitchell and Esther M. Arkin)
- Role: Principal Investigator, Type of Grant: Bilkent University Faculty Research Development Grant, Funding Agency: Bilkent University, Title: Investigations in Geometric Optimization, Amount: US$1,500, Duration: January 1, 2007 -- December 31, 2007.
- Role: Principal Investigator, Type of Grant: TUBITAK (Turkish Scientific and Technological Research Council) 1002, Funding Agency: TUBITAK, Title: Development of Specific and Efficient Algorithms for Large-Scale Geometric Optimization Problem, Amount: 20,250 YTL, Duration: October 1, 2007 -- October 1, 2008, Project Number: 107M411.
PUBLICATIONS
PhD Thesis
- An interior-point perspective on sensitivity
analysis in
linear programming and semidefinite programming, Ph.D. thesis
submitted to Cornell University. (2001) pdf
Journal
Papers
Please
send
me an e-mail
if you would
like
an electronic copy.
- Sensitivity
analysis in linear
programming and semidefinite
programming using interior-point methods, E. Alper
Yıldırım and Michael
J.
Todd. Mathematical
Programming, 90 (2) pp. 229-261
(2001). DOI
- An interior-point approach to sensitivity analysis
in
degenerate linear programs, E. Alper Yıldırım and Michael
J.
Todd. SIAM
Journal on
Optimization, 12 (3) pp. 692-714
(2002). DOI
- Warm-start strategies in interior-point methods
for linear
programming, E. Alper Yıldırım and Stephen J. Wright.
SIAM
Journal on
Optimization, 12 (3) pp. 782-810
(2002). DOI
- An interior-point perspective on sensitivity
analysis in semidefinite programming, E.
Alper Yıldırım. Mathematics of
Operations
Research,
28 (4) pp. 649-676
(2003). DOI
- Approximate minimum enclosing balls in high
dimensions using
core-sets, Piyush
Kumar, Joseph S. B.
Mitchell,
and E. Alper Yıldırım, The
ACM
Journal
of Experimental Algorithmics, Vol. 8,
Article
1 (2003). (Special
issue devoted to selected papers from the Fifth Workshop
on
Algorithm Engineering and Experiments (ALENEX03))
- Unifying optimal partition approach to sensitivity
analysis in
conic optimization, E. Alper Yıldırım. Journal
of
Optimization Theory and Applications, 122 (2) pp. 405-423
(2004). DOI
- Minimum volume enclosing ellipsoids and core sets, Piyush
Kumar and
E. Alper Yıldırım. Journal
of
Optimization Theory and Applications, 126 (1) pp. 1-21 (2005). DOI Erratum
- On extracting maximum stable sets in perfect
graphs using
Lovász's theta function, E.
Alper Yıldırım and Xiaofei
Fan-Orzechowski. Computational
Optimization and Applications, 33 (2-3) pp. 229-247
(2006). DOI
- On the
minimum volume covering
ellipsoid of ellipsoids, E.
Alper Yıldırım. SIAM
Journal on
Optimization, 17 (3) pp. 621-641
(2006). DOI (Winner of the 2006 INFORMS Optimization Society Young
Researcher Prize)
- On Khachiyan's algorithm for the computation of
minimum-volume
enclosing ellipsoids, Michael
J.
Todd
and E. Alper Yıldırım. Discrete and Applied Mathematics, 155 (13) pp. 1731-1744 (2007). DOI
- Computing
minimum volume enclosing axis-aligned ellipsoids, Piyush
Kumar and
E. Alper Yıldırım. Journal
of
Optimization Theory and Applications, 136 (2) pp. 211-228 (2008). DOI
- Implementation
of warm-start
strategies in interior-point methods for linear programming in fixed
dimension, Elizabeth John and E. Alper Yıldırım.
To appear in Computational
Optimization and Applications. DOI (optimization-online)
- Two
algorithms for the minimum enclosing ball problem,
E. Alper Yıldırım. To appear in SIAM Journal on Optimization. (optimization-online)
Refereed
Conference Proceedings
- Computing core-sets and approximate smallest
enclosing
hyperspheres in high dimensions, Piyush Kumar,
Joseph S. B.
Mitchell,
and E. Alper Yıldırım. Proceedings of the Fifth
Workshop
on
Algorithm Engineering and Experiments (ALENEX03), pp. 45 - 55.
(2003)
Submitted Papers
- New lower bounds on the stability number of a graph, E. Alper Yıldırım. Submitted, June 2007. (optimization-online)
- An algorithm and a core set result for the weighted Euclidean one-center problem, Piyush
Kumar and
E. Alper Yıldırım. Submitted, February 2008. (optimization-online)
THESES
SUPERVISED
Ph.D. Theses
- An
Implementation of Warm-Start
Strategies in Interior-Point
Methods for Linear Programming, Elizabeth
John, Ph.D. in Applied
Mathematics and Statistics, Stony Brook University, August 2005.
- Applications
of Lovász's
Theta and Lagrangian Functions to Certain Deterministic and Stochastic
Optimization Problems, Xiaofei
Fan-Orzechowski, Ph.D. in Applied
Mathematics and Statistics, Stony Brook University, December 2005.
(co-advised with Eugene Feinberg)
SOFTWARE
I participated in the
development of the following
software:
- Minimum
Enclosing Ball:
A MATLAB code that computes the minimum enclosing ball of a given set
of points. The code starts by picking two points and iteratively adds
one point at a time until the desired accuracy is achieved. Each
subproblem is solved by the SOCP solver of SDPT3.
The code
can solve instances of the problem with up to 100,000 points in 1,000
dimensions
on a moderate personal computer. Joint work with Piyush Kumar
and Joseph
Mitchell.
- Maximum
Stable Sets in Perfect
Graphs: A MATLAB code that extracts a maximum stable set
in a
perfect graph using Lovász's theta function on successively
smaller
graphs. Joint work with Xiaofei
Fan-Orzechowski. Available upon request.
- Warm-Start
Strategies in
Interior-Point Methods for Linear
Programming: A modification of the interior-point solver PCx
that
implements warm-start strategies in interior-point methods in linear
programming in fixed dimension. Joint with Elizabeth John. Available
upon request.
TALKS (in
reverse chronological order)
- Efficient Algorithms for Large-Scale Optimization Problems Based on Core Sets (in Turkish), National Conference on Operations Research and Industrial Engineering (YA/EM), Istanbul, July 2008.
- Recent Progress in Core Sets, FoCM (Foundations of Computational Mathematics) Conference, Hong Kong, China, June 2008. (invited talk, session chair)
- Two algorithms for the minimum enclosing ball problem,
SIAM
Conference on Optimization, Boston, May 2008. (minisyposium organizer and chair)
- Two algorithms for the minimum enclosing ball problem,
Institute of Applied Mathematics, General Seminar, Middle East
Technical University (METU), October 2007. (invited talk)
- On Khachiyan's algorithm for the computation of minimum-volume enclosing ellipsoids,
Second Mathematical Programming Society International Conference on
Continuous Optimization, McMaster University, Hamilton, Canada, August
2007. (invited talk, session chair)
- On Khachiyan's algorithm for the computation of minimum-volume enclosing ellipsoids,
22nd European Conference on Operational Research (EURO XXII), Prague,
July 2007. (stream organizer, session organizer, and session chair)
- Two algorithms for the minimum enclosing ball problem,
Joint EUROPT-OMS Meeting, Prague, July 2007. (member of the organizing
committee, session organizer, and session chair)
- On the
minimum volume covering ellipsoid of ellipsoids, 2006
INFORMS Annual Meeting, Pittsburgh, November 2006. (invited talk for
the INFORMS Optimization Society Award Session)
- On the
minimum volume covering ellipsoid of ellipsoids,
International Symposium on Mathematical
Programming (ISMP) 2006, Rio de Janeiro, August 2006. (cluster
organizer, session organizer, and session chair)
- An
implementation of warm-start strategies in interior-point methods for
linear programming,
Industrial Engineering,
Department Seminar, Bilkent University, March 2006.
- An
implementation of warm-start strategies in interior-point methods for
linear programming,
Institute of Applied Mathematics, General Seminar, Middle East
Technical University (METU), March 2006. (invited talk)
- On extracting
large stable sets
from Lovász's theta function, Operations
and Information
Systems Group Seminar Series, Koç University, March 2006.
(invited talk)
- On extracting
large stable sets
from Lovász's theta function, Industrial
Engineering,
Department Seminar, Bilkent University, September 2005.
- On the
minimum volume covering ellipsoid of ellipsoids,
Continuous Optimization Seminar,
University of Waterloo, Waterloo, Canada, June 2005. (invited talk)
- On the
minimum volume covering ellipsoid of ellipsoids, Workshop
on Mathematical Programming in
Data Mining and Machine Learning, McMaster University, Hamilton,
Canada, June 2005. (session organizer and chair)
- An
implementation of warm-start strategies in interior-point methods for
linear programming,
SIAM Conference on Optimization, Stockholm, May 2005. (minisymposium
organizer and chair)
- On extracting
large stable sets
from Lovász's theta function, Stony
Brook University,
Operations Research Seminar, Stony Brook, February 2005. (invited talk)
- Efficient
algorithms for
large-scale geometric optimization, INFORMS Denver,
October
2004. (invited session organizer and
session chair)
- Identifying
core sets and its
algorithmic implications in large-scale
geometric optimization, New York University, New York,
October
2004. (invited talk)
- Efficient
algorithms for
large-scale geometric optimization problems and core sets,
McMaster University, Canada, May 2004. (invited talk)
- Minimum enclosing balls and ellipsoids for large
data sets, INFORMS Atlanta, October 2003. (invited session
organizer and
session chair)
- Approximate minimum volume enclosing ellipsoids
using core sets,
DIMACS Workshop on Geometric Optimization, Rutgers University, New
Brunswick, May 2003. (invited talk)
- Computing core-sets and approximate smallest
enclosing
hyperspheres in high dimensions, Lehigh University,
Bethlehem,
April 2003. (invited talk)
- Computing core-sets and approximate smallest
enclosing
hyperspheres in high dimensions, IBM T. J. Watson Research
Center,
Yorktown Heights, March 2003. (invited talk)
- Computing core-sets and approximate smallest
enclosing
hyperspheres in high dimensions, Carnegie Mellon University,
Pittsburgh, January 2003. (invited talk)
- An interior-point perspective on sensitivity
analysis in
semidefinite programming, FoCM Conference, Minneapolis,
August
2002. (invited talk)
- On sensitivity analysis in conic programming, SIAM
Conference on Optimization, Toronto, May 2002. (contributed talk)
- On sensitivity analysis in conic programming, University
of Michigan, Ann Arbor, March 2002. (invited talk)
- An interior-point perspective on sensitivity
analysis in
semidefinite programming, INFORMS Miami Beach, November
2001.
(contributed talk)
- An interior-point perspective on sensitivity
analysis in
linear and semidefinite programming, Stanford University,
Stanford, March 2001. (invited talk)
- An interior-point perspective on sensitivity
analysis in
linear and semidefinite programming, University of
Florida, Gainesville, February 2001. (invited talk)
- An interior-point perspective on sensitivity
analysis in
linear and semidefinite programming, Stony Brook University,
Stony Brook, January 2001. (invited talk)
- An interior-point perspective
on sensitivity analysis in
linear and semidefinite programming, McMaster University,
Hamilton, Canada, January 2001. (invited talk)
- An interior-point perspective on sensitivity
analysis in
linear and semidefinite programming, University of
Waterloo, Waterloo, Canada, January 2001. (invited talk)
- An interior-point perspective on sensitivity
analysis in
linear and semidefinite programming, Bilkent University,
January 2001. (invited talk)
- An interior-point perspective on sensitivity
analysis,
INFORMS San Antonio, November 2000. (invited talk)
- An interior-point perspective on sensitivity
analysis in
linear programming, International Symposium on Mathematical
Programming (ISMP) 2000, Atlanta, August 2000. (invited talk)
- Warm-start strategies in interior-point methods
for linear
programming, SIAM Annual Meeting, Puerto Rico, July 2000.
(contributed talk)
- Sensitivity analysis in linear programming and
semidefinite
programming using interior-point methods, INFORMS
Philadelphia,
November 1999. (contributed talk)
- Semidefinite programming and sensitivity
analysis for dummies,
Argonne National Laboratory, Mathematics and Computer Science Division,
July 1999. (invited talk)
RESEARCH
RELATED LINKS
Last
updated on February 6, 2008. Back to my
homepage.