|
|
|
Markov Decision Processes (MDP) Toolbox
NEWS: Version 3.0 (Sep. 2009) is available for MATLAB and Scilab.
CONTENTS
The MDP toolbox proposes functions related to the resolution of
discrete-time Markov Decision Processes: backwards induction, value iteration,
policy iteration, linear programming algorithms with some variants.
The functions were developped with
MATLAB (note that one of the functions requires the
Mathworks Optimization Toolbox and is not available in Scilab)
by the decision team of the
Biometry and Artificial Intelligence Unit of
INRA Toulouse (France).
The version 3.0 (September 2009) adds several functions related to Reinforcement Learning and improves the handling of sparse matrices. For more detail see the README file.
ABOUT MARKOV DECISION PROCESSES
Markov Decision Processes, Martin L. Puterman, John Wiley&Sons, New-York, 1994.
Processus décisionnels de Markov en intelligence artificielle (Volume 1 et 2), sous la direction de O. Sigaud et O. Buffet , Lavoisier, 2008. (French)
TOOLBOX DOCUMENTATION
A documentation is provided with the toolbox as a set of HTML pages which detail each function.
An example modeling a car race is also available for MATLAB. Download the example zip or gz file on the Project Forge Files page.
DOWNLOAD
The toolbox exists for MATLAB and Scilab environments.
The toolbox has been zipped into a file (.zip file).
This file contains the functions, the HTML documentation and a README file.
An alternative UNIX version (.gz file) can also be downloaded.
All files are available on the Project Forge Files page.
To unpack the .zip file, with WINDOWS, use Winzip software, with Unix, use the command 'unzip MDPtoolbox.zip'.
To unpack the .gz file, with Unix, use the command 'tar xzvf MDPtoolbox.tar.gz'.
URL: http://www.inra.fr/bia/T/MDPtoolbox/MDPtoolbox.html
Page created on July 31, 2001. Last update on September 24, 2009.