Ali Kemal Sinop
School of Mathematics,
Institute for Advanced Study,
010 Simonyi Hall,
1 Einstein Dr,
Princeton, NJ 08540.
I am currently a postdoctoral researcher in theoretical computer
science at the
Institute for Advanced
Computational Intractibility, Princeton University. I received my
Carnegie Mellon University, Computer
Science Department where I was fortunate to be advised
by Prof. Venkatesan
My research interests are in theoretical computer science. More
specifically I am interested in:
Prior to coming to CMU, I worked
Corporate Research for two years under supervision
of Dr. Leo Grady
doing research on graph based algorithms for image segmentation.
I received my M.S. degree
from University of
Pennsylvania in Philadelphia, Pennsylvania in Computer and
Information Science and my B.S. from
- hardness of approximation,
- approximation algorithms,
- spectral graph theory and linear algebra.
Last but not least, my hometown
Dere in city
Teaching I was a teaching assistant for the following
Recent Publications (2009-present)
Towards a better approximation for Sparsest Cut?
Approximating Non-Uniform Sparsest Cut via Generalized
Faster SDP Hierarchy Solvers for Local Rounding
Constant Factor Lasserre Gaps for Graph Partitioning
Optimal Column-Based Low-Rank Matrix Reconstruction.
Lasserre Hierarchy, Higher Eigenvalues, and
Approximation Schemes for Quadratic Integer
Programming with PSD Objectives.
The complexity of finding independent sets in bounded degree
(hyper)graphs of low chromatic number.
Improved inapproximability results for maximum k-colorable
Combinatorial Preconditioners and Multilevel Solvers for
Problems in Computer Vision and Image Processing.
Not So Recent Publications
Fast Approximate Random Walker Segmentation Using Eigenvector
A Seeded Image Segmentation Framework Unifying Graph Cuts and
Random Walker Which Yields A New Algorithm.
Uninitialized, Globally Optimal, Graph-Based Rectilinear
Shape Segmentation - The Opposing Metrics Method.
Accurate Banded Graph Cut Segmentation of Thin Structures
Using Laplacian Pyramids.