Hierarchical automatic differentiation by vertex elimination and source transformation.
Date
2003-01-01T00:00:00Z
Supervisor/s
Journal Title
Journal ISSN
Volume Title
Publisher
Springer-verlag
Department
Type
Book chapter
ISSN
0302-9743
Format
Free to read from
Citation
Mohamed Tadjouddine, Shaun A. Forth & John D. Pryce, Hierarchical Automatic
Differentiation by Vertex Elimination and Source Transformation. Computational
Science and Its Applications - ICCSA 2003, International Conference, Montreal,
Canada, May 18-21, 2003, Proceedings, Part II, Lecture Notes in Computer Science
, Vol. 2668, pp115-124, Kumar, V.; Gavrilova, M.L.; Tan, C.J.K.; L'Ecuyer, P.
(Eds.)
Abstract
We present a hierarchical scheme to extend the applicability of automatic differentiation (AD) by vertex elimination from the basic block level to code with branches and subroutine calls. We introduce the ELIAD tool that implements our scheme. Results from computational fluid dynamics (CFD) flux linearisations show runtime speedup by a typical factor of two over both finite-differencing and traditional forward and reverse modes of AD.