MATH Seminar

Title: Gradient Descent Methods for Large-Scale Linear Inverse Problems
Defense: Honors thesis
Speaker: Chen (Cool) Cheng of Emory University
Contact: James Nagy, nagy@mathcs.emory.edu
Date: 2013-04-11 at 4:00PM
Venue: W306
Download Flyer
Abstract: Iterative gradient descent methods are frequently used for ill-posed inverse problems because they are suitable for large models and they are cheap to work with. In this thesis, we explore three different types of gradient descent methods: the Landweber method, method of steepest descent, and the Barzilai-Borwein method. Specifically, we also compare the efficiency of these methods to the conjugate gradient method. The thesis begins with an introduction to the history and application of the gradient descent methods and to the methods tested, and follows with convergence analysis and numerical experience on real images. Ways to accelerate and smooth the BB method are also included.