Friday, July 18

10:30 AM-12:30 PM
Terman Auditorium

Verified Methods with Computational Differentiation

The purpose of this minisymposium is to explore the overlap between the methods of verified computation and the methods of computational differentiation. In the past, both fields have stimulated each other repeatedly, and both have borrowed techniques and tools from the other. The speakers in this minisymposium will provide an overview over the two fields, and then focus on new and future overlap of the approaches.

Organizer: Martin Berz
Michigan State University

10:30 Computational Differentiation in Global Optimization Software
George Corliss, Marquette University; and R. Baker Kearfott, University of Southwestern Louisiana
11:00 Computational Differentiation Methods for Nonsmooth Problems
Andreas Griewank, Technische Universitat Dresden, Germany
11:30 Taylor Models and their Applications
Martin Berz, Organizer
12:00 Verified Analysis of Stability in Particle Accelerators
Kyoko Makino, Michigan State University
12:30 Computational Complexity of Interval Computations, of Taylor Methods, and of Their Fractal Extensions
Vladik Kreinovich and Brian Penn, University of Texas, El Paso; and John W. Harbaugh, Stanford University

AN97 Homepage | Program Updates|
Registration | Hotel and Dormitory Information | Transportation | Program-at-a-Glance | Program Overview

MMD, 6/6/97