Dr. Jill Reese will attend the American Institute of Mathematics workshop on Derivative-Free Hybrid Optimization Methods for Solving Simulation-Based Problems in Hydrology, October 6-10, 2008. The workshop participants will form an information commons to create and exchange ideas on applying hybrid optimization methods to hydrology problems which exhibit behaviors that often challenge traditional, derivative-based optimization algorithms. The participants will suggest problems and questions at the begining of the 5-day workshop, then split into organized subgroups and discuss issues such as water quality, water supply, plume containment, and drainage basin management. Subgroups will then reconvene with the other participants to discuss the concepts and research possibilities examined in smaller group discussions. The workshop, sponsored jointly by AIM and the National Science Foundation, will bring together an intellectual and academic mix of approximately 20 new faculty, research scholars, graduate students, and seasoned professors. Some specific areas of interest include:

  • development and analysis of hybrid optimization methods to improve accuracy, efficiency and reliability of solutions for this class of problems and to exploit the strengths of a variety of derivative-free optimization and sampling methods not commonly used in hydrology,
  • understanding how to choose appropriate algorithms for a specific type of problem, which involves a classification of both problem types and hybrid approaches,
  • developing metrics to compare hybrid algorithms, and study these comparisons on benchmarking problems from hydrology.

Workshop lectures will familiarize the participants with the background material leading up to specific problems before parallel working sessions begin. For more information please click here.

Dept. of Scientific Computing
Florida State University
400 Dirac Science Library
Tallahassee, FL 32306-4120
Phone: (850) 644-1010
admin@sc.fsu.edu
© Scientific Computing, Florida State University
Scientific Computing