|
Organizers |
Development of Tool- and Language-Independent Automatic Differentiation Modules: Current Work and Future Directions
by
Jason Abate
Texas Institute for Computational and Applied Mathematcs, The University of Texas at Austin
In recent years, we have studied the development of AD augmentation modules that operate independent of specific AD tools and languages. We describe our motivations for this work and initial experiments in developing a polyalgorithmic Hessian module which plugs into two existing tools, which we have found to be a valuable approach to building AD tools. Finally we give some thoughts based on this experience as to how this process could be improved.
Date received: December 30, 1999
Copyright © 1999 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cads-37.