Michael Todd

Todd was educated at the University of Cambridge and Yale University. He taught briefly at the University of Ottawa before moving to Cornell University, where he worked for forty-one years. His research covers fixed-point algorithms, and simplex, ellipsoid, and interior-point methods for linear, nonlinear, and conic programming. He is a fellow of SIAM and of INFORMS, and a member of the National Academy of Engineering.

Title of talk: The Ellipsoid Method for Linear Inequalities

In 1979, the West learned of Khachiyan's polynomial-time algorithm for linear programming based on the ellipsoid method. That fall, Don Goldfarb came to Cornell for a year, and we wrote two papers on the subject (one also with Bob Bland). I will describe that work and later contributions to the subject, including connections to interior-point methods.