Publication: A benchmark of selected algorithmic differentiation tools on some problems in computer vision and machine learning
Introduction
Applications
Tools
Research Groups
Workshops
Publications
   List Publications
   Advanced Search
   Info
   Add Publications
My Account
About

A benchmark of selected algorithmic differentiation tools on some problems in computer vision and machine learning

- Article in a journal -
 

Author(s)
Filip Srajer , Zuzana Kukelova , Andrew Fitzgibbon

Published in
Special issue of Optimization Methods & Software: Advances in Algorithmic Differentiation Optimization Methods & Software

Editor(s)
Bruce Christianson, Shaun A. Forth, Andreas Griewank

Year
2018

Publisher
Taylor & Francis

Abstract
Algorithmic differentiation (ad) allows exact computation of derivatives given only an implementation of an objective function. Although many ad tools are available, a proper and efficient implementation of ad methods is not straightforward. The existing tools are often too different to allow for a general test suite. In this paper, we compare 15 ways of computing derivatives including 11 automatic differentiation tools implementing various methods and written in various languages (C++, F#, MATLAB, Julia and Python), 2 symbolic differentiation tools, finite differences and hand-derived computation.We look at three objective functions from computer vision and machine learning. These objectives are for the most part simple, in the sense that no iterative loops are involved, and conditional statements are encapsulated in functions such as abs or logsumexp. However, it is important for the success of ad that such ‘simple’ objective functions are handled efficiently, as so many problems in computer vision and machine learning are of this form.

Cross-References
Christianson2018Sio

BibTeX
@ARTICLE{
         Srajer2018Abo,
       author = "Filip Srajer and Zuzana Kukelova and Andrew Fitzgibbon",
       title = "A benchmark of selected algorithmic differentiation tools on some problems in computer
         vision and machine learning",
       journal = "Optimization Methods \& Software",
       xvolume = "0",
       xnumber = "0",
       pages = "889--906",
       year = "2018",
       publisher = "Taylor \& Francis",
       doi = "10.1080/10556788.2018.1435651",
       url = "https://doi.org/10.1080/10556788.2018.1435651",
       eprint = "https://doi.org/10.1080/10556788.2018.1435651",
       crossref = "Christianson2018Sio",
       volume = "33",
       number = "4--6",
       abstract = "Algorithmic differentiation (AD) allows exact computation of derivatives given only
         an implementation of an objective function. Although many AD tools are available, a proper and
         efficient implementation of AD methods is not straightforward. The existing tools are often too
         different to allow for a general test suite. In this paper, we compare 15 ways of computing
         derivatives including 11 automatic differentiation tools implementing various methods and written in
         various languages (C++, F\#, MATLAB, Julia and Python), 2 symbolic differentiation tools,
         finite differences and hand-derived computation.We look at three objective functions from computer
         vision and machine learning. These objectives are for the most part simple, in the sense that no
         iterative loops are involved, and conditional statements are encapsulated in functions such as abs
         or logsumexp. However, it is important for the success of AD that such ‘simple’
         objective functions are handled efficiently, as so many problems in computer vision and machine
         learning are of this form.",
       booktitle = "Special issue of Optimization Methods \& Software: Advances in
         Algorithmic Differentiation",
       editor = "Bruce Christianson and Shaun A. Forth and Andreas Griewank"
}


back
  

Contact:
autodiff.org
Username:
Password:
(lost password)