2 Matching Results

Search Results

Advanced search parameters have been applied.

Reducing the memory requirement in reverse mode automatic differentiation by solving TBR flow equations.

Description: The fast computation of gradients in reverse mode Automatic Differentiation (AD) requires the generation of adjoint versions of every statement in the original code. Due to the resulting reversal of the control flow certain intermediate values have to be made available in reverse order to compute the local partial derivatives. This can be achieved by storing these values or by recomputing them when they become required. In any case one is interested in minimizing the size of this set. Following an extensive introduction of the ''To-Be-Recorded'' (TBR) problem the authors present flow equations for propagating the TBR status of variables in the context of reverse mode AD of structured programs.
Date: January 11, 2002
Creator: Naumann, U.
Partner: UNT Libraries Government Documents Department

An introduction to using software tools for automatic differentiation.

Description: The authors give a gentle introduction to using various software tools for Automatic Differentiation (AD). Ready-to-use examples are discussed and links to further information are presented. The target audience includes all those who are looking for a straight-forward way to get started using the available AD technology. The document is supposed to be dynamic in the sense that its content will be kept up-to-date as the AD software covered is evolving.
Date: October 7, 2003
Creator: Naumann, U. & Walther, A.
Partner: UNT Libraries Government Documents Department