Source Transformation for MATLAB Automatic Differentiation.
Date
2005-12
Authors
Supervisor/s
Journal Title
Journal ISSN
Volume Title
Publisher
Springer-Verlag
Department
Type
Postprint
ISSN
Format
Free to read from
Citation
Rahul V. Kharche & Shaun A. Forth, Source Transformation for MATLAB Automatic Differentiation, Computational Science - ICCS 2006 6th International Conference, Reading, UK, May 28-31, 2006, Proceedings, Part IV, pages 558-565.
Editors Vassil Alexandrov, Dick van Albada, Peter Sloot and Jack Dongarra, Lecture Notes in Computer Science 3994,
Abstract
We present MSAD, a source transformation implementation of forward mode automatic differentiation for MATLAB. MSAD specialises and inlines operations from the fmad and derivvec classes of the MAD package. The operator overloading overheads inherent in MAD are eliminated while preserving the derivvec class's optimised derivative combination operations. Compared to MAD, results from several test cases demonstrate significant improvement in efficiency across all problem sizes.