The Interior-Point Revolution in Optimization
Margaret H. Wright
Usage of the word ``optimization'', once a technical term, has grown during the past twenty years to cover essentially every aspect of life, including dental hygiene, hair care, and pop psychology. During the same period, the classical field of constrained continuous optimization---finding the best value of an objective subject to constraints---has not been idle: a sweeping change, sometimes called the ``interior-point revolution'', started in 1984 with Karmarkar's announcement of a polynomial-time algorithm for linear programming that was also fast in practice. Since then, interior-point methods have continued to provide a fascinating mixture of old and new ideas, with implications to topics ranging from linear programming to approximation algorithms for NP-hard problems. This talk will describe the evolution of interior methods from 1984 to the present, touching on controversies and puzzles (some resolved, some not) and directions for future research.
Short Biography:
Margaret H. Wright is Silver Professor of Computer Science and chair of the Computer Science Department at New York University's Courant Institute of Mathematical Sciences. She received her B.S. in Mathematics, and her M.S. and Ph.D. in Computer Science, from Stanford University. From 1988-2001 she worked in the Computing Science Research Center at Bell Laboratories, where she was named a Bell Labs Fellow in 1999.
Her research interests include optimization, linear algebra, numerical and scientific computing, and scientific and engineering applications.
She is a member of the National Academy of Engineering and the American Academy of Arts and Sciences. From 1995-1996 she served as president of the Society for Industrial and Applied Mathematics (SIAM). She currently chairs the Advisory Committee on Advanced Scientific Computing for the Department of Energy's Office of Science, and she was a member of the Blue Ribbon Panel on Cyberinfrastructure for the National Science Foundation.
Wright is Editor-in-Chief of SIAM Review, as well as an associate editor of the SIAM Journal on Scientific Computing, the SIAM Journal on Optimization, and Mathematical Programming.
Homepage: http://www.nyu.edu/fas/Faculty/WrightMargaret.html