An Adaptive Preconditioned Nonlinear Conjugate Gradient Method with Limited Memory

	Computational Mathematics Seminar Series
	An Adaptive Preconditioned Nonlinear Conjugate Gradient Method with Limited Memory
	Hongchao Zhang, LSU
	Assistant Professor, Department of Mathematics
	Johnston Hall 338 November 29, 2011 - 03:30 pm
Abstract: Nonlinear conjugate gradient methods are an important class of methods for solving large-scale unconstrained nonlinear optimization. However, their performance is often severely affected when the problem is very ill-conditioned. In the talk, efficient techniques for adaptively preconditioning the nonlinear conjugate method in the subspace spanned by a small number of previous searching directions will be discussed. The new method could take advantages of both nonlinear conjugate methods and limited-memory BFGS quasi-Newton methods, and achieves significant performance improvement compared with CG\_DESCENT conjugate gradient method and L-BFGS quasi-Newton method.

Speaker's Bio: Hongchao Zhang is an Assistant Professor jointly in the Mathematics Department and the Center for Computation & Technology (CCT) at Louisiana State University. He received a Ph.D. degree in the Department of Mathematics from the University of Florida. His research interests are nonlinear optimization theory, algorithms and application.

Refreshments will be served.
This lecture has a reception.

CENTER FOR COMPUTATION & TECHNOLOGY