TU Darmstadt / ULB / TUbiblio

Source Transformation for the Optimized Utilization of the Matlab Runtime System for Automatic Differentiation

Hück, Alexander and Willkomm, J. and Bischof, Christian
Mehl, Miriam and Bischoff, Manfred and Schäfer, Michael (eds.) (2015):
Source Transformation for the Optimized Utilization of the Matlab Runtime System for Automatic Differentiation.
In: Recent Trends in Computational Engineering - CE2014, Springer, pp. 115-131, [Online-Edition: https://www.springer.com/de/book/9783319229966#],
[Book Section]

Abstract

Operator overloading in Matlab allows for user-defined types to semantically augment existing Matlab codes without changes. However, given sufficient knowledge about types and operand semantics, operator overloading can be replaced by equivalent function calls. The ADiMat software tool provides automatic differentiation of Matlab functions using a hybrid approach that combines source transformation and operator overloading. It can also be used as a general framework for user-defined transformations of Matlab codes. Tests showed the potential for performance improvement in a derivative class providing essential linear algebra functionality for ADiMat. The Matlab runtime environment was benchmarked regarding function and method call overheads as well as property access overhead with Matlab's objects. These tests identify the cell class property access as the main performance culprit. Hence, an automatic method, transforming the XML based abstract syntax tree created with ADiMat's toolchain through a set of stylesheets, was developed. This process completely removes the derivative object usage and hence the opreator overloading and the property access overhead from any derivative function created with ADiMat. Experimental results show that performance is improved considerably depending on the data container storing the derivative directions.

Item Type: Book Section
Erschienen: 2015
Editors: Mehl, Miriam and Bischoff, Manfred and Schäfer, Michael
Creators: Hück, Alexander and Willkomm, J. and Bischof, Christian
Title: Source Transformation for the Optimized Utilization of the Matlab Runtime System for Automatic Differentiation
Language: English
Abstract:

Operator overloading in Matlab allows for user-defined types to semantically augment existing Matlab codes without changes. However, given sufficient knowledge about types and operand semantics, operator overloading can be replaced by equivalent function calls. The ADiMat software tool provides automatic differentiation of Matlab functions using a hybrid approach that combines source transformation and operator overloading. It can also be used as a general framework for user-defined transformations of Matlab codes. Tests showed the potential for performance improvement in a derivative class providing essential linear algebra functionality for ADiMat. The Matlab runtime environment was benchmarked regarding function and method call overheads as well as property access overhead with Matlab's objects. These tests identify the cell class property access as the main performance culprit. Hence, an automatic method, transforming the XML based abstract syntax tree created with ADiMat's toolchain through a set of stylesheets, was developed. This process completely removes the derivative object usage and hence the opreator overloading and the property access overhead from any derivative function created with ADiMat. Experimental results show that performance is improved considerably depending on the data container storing the derivative directions.

Journal or Publication Title: Recent Trends in Computational Engineering - CE2014
Title of Book: Recent Trends in Computational Engineering - CE2014
Series Name: Lecture Notes in Computational Science and Engineering
Volume: 105
Publisher: Springer
ISBN: 978-3-319-22996-6
Uncontrolled Keywords: Source Transformation, Automatic Differentiation, ADiMat, Matlab, Performance Measurement, XSLT, XML AST
Divisions: 20 Department of Computer Science
20 Department of Computer Science > Scientific Computing
Zentrale Einrichtungen
Zentrale Einrichtungen > University IT-Service and Computing Centre (HRZ)
Zentrale Einrichtungen > University IT-Service and Computing Centre (HRZ) > Hochleistungsrechner
Date Deposited: 14 Jul 2015 08:29
Official URL: https://www.springer.com/de/book/9783319229966#
Identification Number: doi:10.1007/978-3-319-22997-3_7
Export:

Optionen (nur für Redakteure)

View Item View Item