AMOR

 

Hierarchical Automatic Differentiation by Vertex Elimination and Source Transformation

Mohamed Tadjouddine, Shaun A Forth  & John D. Pryce 

Published in 
COMPUTATIONAL SCIENCE AND ITS APPLICATIONS - ICCSA 2003, PT 2, PROCEEDINGS
LECTURE NOTES IN COMPUTER SCIENCE 2668, SPRINGER-VERLAG BERLIN, BERLIN
115-124, 2003

Presented at
International Conference on Computational Science and Its Applications (ICCSA 2003)
MONTREAL, CANADA, MAY 18-21, 2003


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.

Download
Authors' uncorrected PDF:
mt_iccs03.pdf (0.4 MB) 
Publisher's Web Entry:  http://www.springerlink.com/openurl.asp?genre=article&issn=0302-9743&volume=2668&spage=115

AMOR home