Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Topics - Anders L Madsen

Pages: [1] 2 3 ... 6
1
AMIDST / Project Description
« on: August 11, 2014, 10:35:15  »
The web-site http://amidst.hugin.com is dedicated to hosting information on the AMIDST project with the main focus on the use of Bayesian networks.

2
Probabilistic Graphical Models (PGM2014) / Call for papers
« on: February 06, 2014, 15:36:21  »
By request of Silja Renooij, Universiteit Utrecht we post this CfP:

Call for Papers
 PGM 2014: The Seventh European Workshop on Probabilistic Graphical Models
 September 17-19, 2014
 Utrecht, The Netherlands
 __________________________________________________________________________

 Important Dates:

 Deadline for abstract submissions: May 12, 2014
 Deadline for paper submissions: May 16, 2014
 Notification of acceptance: June 23, 2014
 Final versions due: July 18, 2014
 Workshop dates September 17-19, 2014

 __________________________________________________

 Call for papers:

The aim of the workshop is to bring together people interested in
probabilistic graphical models and to provide a forum for discussion
of the latest research developments in this field. To promote interactions
among the participants, parallel sessions will be avoided and the workshop
will be organised around a single thread of plenary and poster sessions.
We welcome theoretical and applied contributions related to various aspects
of probabilistic graphical models, such as:

 - Principles of Bayesian (belief) networks, chain graphs, decision networks, influence diagrams,
   probabilistic relational models, and other probabilistic graphical models (PGMs).
 - Information processing in PGMs (exact and approximate inference).
 - Learning and data mining in the context of PGMs: machine learning approaches,
   statistical testing and search methods, MCMC simulation.
 - Exploitation for the construction of PGMs of results from related disciplines
   such as statistics, information theory, optimization, and decision making under uncertainty.
 - Software systems based on PGMs.
 - Application of PGMs to real-world problems.

Papers submitted for review should report on original, previously unpublished work.
Each submitted paper will be reviewed by at least three reviewers.


Proceedings:
 
All accepted papers will be included in the workshop proceedings.
At this stage, we are negotiating publication by a renowned publisher.


Submission procedure:

PGM 2014 requires electronic submission of papers according to the
instructions found on the web site of the workshop:

 http://www.projects.science.uu.nl/PGM/

The authors must prepare their papers according to the Springer LNCS
format, and submit the full version of the papers by the deadline stated above.
The page limit is 16 pages, including figures and bibliography.
The web page for author instructions contains further format information
and provides access to style files and templates.


Paper presentation:

Papers will be accepted for plenary or poster presentation.
In the proceedings, no distinction is made between the two.
At least one author of accepted papers is required to register for the workshop.

After the workshop, authors of selected papers will be invited to submit
an extended version of their paper for a special issue of a journal, which is yet to be decided.
The extended papers will undergo a full reviewing process.

Organisation:

The workshop is hosted by the Department of Information and Computing Sciences,
Utrecht University, The Netherlands.

3
OpenNESS / Project Description
« on: January 22, 2014, 19:44:25  »
The web-site http://openness.hugin.com is dedicated to hosting information on the OpenNESS project with the main focus on the use of Bayesian networks.

4
OpenNESS / HUGIN OpenNESS installation package
« on: January 17, 2014, 10:44:13  »
Please find the installation package here: http://openness.hugin.com/gui

5
SelSus / Project Description
« on: November 28, 2013, 11:24:58  »
The web-site http://selsus.hugin.com is dedicated to hosting information on the SelSus project with the main focus on the use of Bayesian networks.

6
OpenNESS / The Bayesian network web site
« on: March 07, 2013, 23:34:01  »
The web-site http://openness.hugin.com is dedicated to hosting information on the OpenNESS project with the main focus on the use of Bayesian networks.

7
FAQ / Why do the number not sum to one?
« on: October 22, 2012, 16:22:03  »
HUGIN uses floating point to support the representation and use of real numbers (e.g., probabilities), see e.g., http://en.wikipedia.org/wiki/Floating_point. On the wiki-page there is useful information on the use of floating point and its limitations.

In floating point a real number is represented using a (limited) number of significant digits. The number of significant digits depends on the data type used to represent a real number, i.e., the precision. In HUGIN we use either single (4 bytes) or double precision (8 bytes).

In single precision you have 7 significant digits. This means that single precision has sufficiently many digits to store, for instance, the number 0.1234567 in a variable with seven significant digits. A number such as 0.99999999 is, on the other hand, rounded to 1 and if you add 0.00000001, the number 0.00000001 is lost due to the limited precision (only seven significant digits).

As an example consider using HUGIN ActiveX server (single precision) in Excel to add the numbers above (i.e, 0.99999999 + 0.00000001). If Excel uses double precision, the sum becomes 1.00000001 (due to the fact that 0.99999999 is rounded to 1 by HUGIN due to the single precision nature of the HUGIN ActiveX server).

There are other examples which cannot be solved. For instance, consider computing the sum 1/3 + 1/3 + 1/3 using floating point in any (finite) precision. This will never sum to one no matter how many digits of precision you have. Thus, in a computer probabilities do not always sum to one.


8
The default package is strictly a per user installation. Only the windows user account that installed HUGIN can run the program.

To install just log on to the target windows user account and double-click the installation file, afterwards only this windows user account can run the program.  (DO NOT right-click-run-as-administrator the installation file - this will result in the program being installed for the administrator Windows user account and not the target Windows user account as desired).
 
The 'admin' package is a machine wide installation. Any windows user account on the computer can run the program. If you do not have the 'admin' installation package and a corresponding 'admin' license key, then please contact sales-at-hugin.com.


9
Probabilistic Graphical Models (PGM2012) / Call for Papers
« on: May 03, 2012, 09:51:40  »
By request of Manuel Gomez Olmedo, Universidad de Granada we post this CfP

Probabilistic Graphical Models (PGM2012) Workshop
Granada, Spain, 19-21 September 2012 (http://leo.ugr.es/pgm2012)

The submission deadline has been extended to June 4th (abstract) and June 8th (full paper)
(see list of important dates below)

Call for papers:

The aim of the workshop is to bring together people interested in probabilistic graphical models
and provide a forum for discussion of the latest research developments in this field. The workshop
is organized so as to facilitate discussions and collaboration among the participants also outside
the workshop sessions. To facilitate the spreading of research results and promote interactions
among the participants, parallel sessions will be avoided and the workshop will be organized
around a single thread of plenary and posters sessions.


We welcome theoretical and applied contributions related to the following topics:

- Principles of Bayesian (belief) networks, chain graphs, decision networks, influence diagrams,
  and other probabilistic graphical models (PGMs).
- Information processing in PGMs (exact and approximate inference).
- Learning and data mining in the context of PGMs: machine learning approaches, statistical testing
  and search methods, MCMC simulation.
- Exploitation of results from related disciplines for the construction of PGMs, e.g., statistics, information
  theory, optimization, decision making under uncertainty, multivariate statistical analysis.
- Software systems based on PGMs.
- Application of PGMs to real-world problems.

We invite submissions that concern oneCHANGED or more of the above topics, or any other aspects related to
probabilistic graphical models. Papers submitted for review should report on original, previously
unpublished work. Each submitted paper will be reviewed by at least two reviewers.

Submission procedure

PGM 2012 requires electronic submission of papers according to the instructions found on the web page
http://leo.ugr.es/pgm2012/author-instructions.php. The authors must prepare their papers according to
the PGM proceedings format, and submit the full version of the papers by the deadline found below. The
page limit is 8 pages, including figures and bibliography.  The page for author instructions contains format
information and provides access to style files and templates.

Paper presentation

Papers will be accepted for plenary or poster presentation. All accepted papers will be included in the
proceedings (with ISBN), which will be made available at the workshop reception. At least one author
of accepted papers is expected to attend the workshop. After the workshop, authors of selected papers
will be invited to submit an extended version of their paper to appear in a special issue of a high quality
journal. All of the submitted papers will undergo a full reviewing process.

Important Dates

Call for papers opened    November 2011
CHANGED: Deadline for abstract submissions    June 4, 23:59 GMT, 2012
CHANGED: Paper submissions    June 8, 23:59 GMT, 2012
CHANGED:  Notification of acceptance    July 20, 2012
CHANGED: Final versions due    August 24,  23:59 GMT, 2012
Welcoming reception    September 18, 2012
Workshop dates    September 19-21, 2012

10
Environmental conditions play a crucial role in the distribution and abundance of fish species in any area, but the relative importance of various environmental factors is very difficult to determine. The coastline of Finland in the northern Baltic Sea offers a unique natural experimental setting that can be used to assess the relative importance of various environmental factors for the species occupying it. The area includes major variations in several crucial environmental factors: salinity, temperature regime, represented by winter ice duration, coastline characteristics, and eutrophic status.

Demonstration of the costal fish production model:
http://demo.hugin.com/index.php/Coastal_Fish_Production

11
UAI 9th Bayesian Modeling Applications Workshop / Call for Papers
« on: April 10, 2012, 18:41:24  »
By request of Prof. Ann Nicholson Monash University we post this CfP


UAI 9th Bayesian Modeling Applications Workshop

                                
Call for Papers
               
Saturday, August 18, 2012, Catalina Island

                
Due to requests, extended submission deadline is Saturday 19th May.

Special theme: Temporal Modeling

The 9th Bayesian Modeling Applications Workshop solicits submissions of real-world applications of graphical models and Bayesian networks, in particular those dealing with temporal modeling.  Our desire is to foster discussion and interchange about novel contributions that can speak to both the academic and the larger research community. Accordingly, we seek submissions also from practitioners and tool developers as well as researchers.

Bayesian networks are now a powerful, well-established technology for reasoning under uncertainty, supported by a wide range of mature academic and commercial software tools. They are now being applied in many domains, including environmental and ecological modeling, bioinformatics, medical decision support, many types of engineering, robotics, military, financial and economic modeling, education, forensics, emergency response, surveillance, and so on. We welcome submissions describing such real world applications, whether asstand-alone BNs or where the BNs are embedded in a larger software system. We encourage authors to address the practical issues involved in developing real-world applications, such as knowledge engineering methodologies, elicitation techniques, defining and meeting client needs, validation processes and integration methods, as well as software tools, including visualization and user interaction techniques to these support these activities.

We particularly encourage the submission of papers that address the workshop theme of temporal modeling.  Recently communities building dynamic Bayes networks (DBNs) and partially observable MDPs (POMDPs) are coming to realize that they are applying their methods to identical applications.  Similarly POMDPs and other probabilistic methods are now established in the field of Automated Planning. Stochastic process models such as continuous time Bayes networks (CTBNs) should also be considered as part of this trend. Adaptive and on-line learning models also fit into this focus.


Submissions
-----------

Submissions should be in UAI format, limited to 8 pages, using the online submission process. All papers will be peer reviewed by at least two independent referees.  Papers must be submitted electronically in PDF format via the EasyChair online submission page:
http://www.easychair.org/conferences/?conf=bmaw12

We encourage co-submission of applications papers that have been submitted to the main UAI 2012 conference, as our submission deadline comes before the UAI acceptance deadline. If accepted for UAI, the paper would be published in UAI proceedings, but we will invite a poster presentation at the Workshop also.

Workshop format
 --------------

The format the workshop will be combination of oral and poster
presentations, with demonstrations encouraged for both, grouped in
sessions to facilitate discussion.


Publication of Proceedings
--------------------------

As in past years, proceedings will be published online (as a CEUR
Workshop Proceedings Volume), and selected papers will be invited to
submit to a special issue of a journal.


Important Dates
---------------

   * 1 May 2012:       Abstract submission
   * 5 May 2012:       Full paper submission
   * 11 June 2012:     Author notification
   * 31 July 2012:      Camera ready copy due
   * 18 August 2012: Workshop (following the UAI2012 main conference, Aug 15-17)


Workshop Chairs
-------------

John Mark Agosta (Toyota ITC, USA)
Ann Nicholson (Monash University & Bayesian Intelligence, Australia)
M. Julia Flores (University of Castilla-La Mancha, Spain)

Program Committee
-----------------

Concha Bielza, Technical University of Madrid, Spain
Dennis M. Buede, Innovative Decisions, Inc., USA
Javier Diez,  UNED, Spain
Marek Druzdzel, University of Pittsburgh, USA & Bialystok University
of Technology, Poland
José A. Gámez, University of Castilla-La Mancha, Spain
Asela Gunawardana, Microsoft Research, USA
Branislav Kveton, Technicolor Research, Palo Alto, USA
Helge Langseth, The Norwegian University of Science and Technology, Norway
Pedro Larrañaga, Technical University of Madrid, Spain
Philippe Leray, Polytech'Nantes, France
Michael Mahoney, Stanford University, Dept of Mathematics, USA
Suzanne Mahoney, Innovative Decisions, USA
Chris Meek,  Microsoft Research, USA
Ole Mengshoel, CMU Silicon Valley Campus, USA
Serafín Moral, University of Granada, Spain
Jens D. Nielsen,  University of Sheffield, UK
Thomas D. Nielsen, Aalborg University, Denmark
Jose M. Peña, Linköping University, Sweden
Pascal Poupart, University of Waterloo, Canada
Antonio Salmerón, University of Almería, Spain
Fabio Stella, University of Milano-Bicocca, Italy
L. Enrique Sucar, INAOE, Mexico
Georgios Theocharous, Yahoo Research, USA

12
Agriculture / Agricultural Risk Management
« on: March 27, 2012, 10:15:46  »
HUGIN is proud to announce its participation in a project on agricultural risk management with focus on pig production.  The project is externally funded by Norma and Frode S. Jacobsens Fond and Nordea Bank Fonden.  The coordinator of the project is Mogens Lund, Institute of Food and Resource Economics, University of Copenhagen.

The main objective of the project is to develop a web-based solution for financial risk management at the farm level using Bayesian network models.

The web page of the project can be found here:http://risiko-svinebrug.hugin.com/ (in Danish)

13
Java / Learning using the Java API
« on: September 23, 2011, 11:10:37  »
Here is an example on learning using the HUGIN Java API
Code: [Select]
/*
 * This simple Java program illustrates how to learn (the parameters)
 * of a Bayesian network from data
 *
 * The result is stored in a file name learning.net
 *
 *
 * Author: Anders L Madsen @ HUGIN EXPERT A/S
 *
 * For any questions or comments, please contact the author at
 * alm@hugin.com
 */
import COM.hugin.HAPI.*;
import java.text.DecimalFormat;

class EnterDataExample {
    // the model
    Domain dom = null;

    // the nodes / variables in the model
    LabelledDCNode Sex, Height;

    public EnterDataExample ()
    {
try {
     // Create domain. Done once.
     dom = new Domain ();
     
    Sex = new LabelledDCNode(dom); Sex.setName ("Sex");
     Height = new LabelledDCNode(dom); Height.setName ("Height");

     Sex.setNumberOfStates (2);
     Sex.setStateLabel (0, "Female");
     Sex.setStateLabel (1, "Male");

     Height.setNumberOfStates (2);
     Height.setStateLabel (0, "Low");
     Height.setStateLabel (1, "High");

     // Add Sex as parent of Height. Comment this line if you
     // uncomment "dom.learnStructure ();" below
     Height.addParent (Sex);

     // create Experience tables to enable CPT estimatio
     Sex.getExperienceTable ();
     Height.getExperienceTable ();

} catch (Exception e) {
     e.printStackTrace ();
     System.err.println (e.getMessage ());      
}
    }

    // Data is an array of strings. First entry is Sex and second is
    // Height;
    String dataArray[][]  =
{
    {"Male", "High"},
     {"Male", "High"},
     {"Male", "High"},
     {"Female", "High"},
     {"Female", "Low"},
     {"Male", "Low"},
     {"Female", "Low"},
     {"Male", "High"},
     {"Male", ""},
     {"Female", "Low"}
};
;
    int N=10;

    // Main procedure. 
    protected void doLearning () {
try{
    // enter data
     enterData ();

     // (learn structure) compile and estimate parameters
     // dom.learnStructure (); // uncomment this line (and the
     // addParent line) above to learn structure from data
     dom.compile ();
     dom.learnTables ();

     // save domain in file
     dom.saveAsNet ("learning.net");
} catch (Exception e) {
     e.printStackTrace ();
     System.err.println (e.getMessage ());      
}
    }

    //
    protected void enterData () {
try{
    long j, s;
     for (int i = 0; i < N; i++) {
j = dom.newCase ();

s = Sex.getStateIndex (dataArray[i][0]);
if (s >= 0)
     Sex.setCaseState (j, s);

s = Height.getStateIndex (dataArray[i][1]);
if (s >= 0)
     Height.setCaseState (j, s);
     }
     //... more data
} catch (Exception e) {
     e.printStackTrace ();
     System.err.println (e.getMessage ());      
}
    }
   
    static public void main (String args[])
    {
EnterDataExample ede = new EnterDataExample ();
ede.doLearning ();
    }
}



14
Patterns & Predictions / End of Life
« on: September 21, 2011, 14:53:00  »
This product has reached end of life.

15
C# / Matlab: How to load an OOBN
« on: July 13, 2011, 18:06:11  »
Using the HUGIN .NET API it is possible to make use of HUGIN functionality in Matlab. The appropriate HUGIN dll file should be loaded as an assembly into Matlab.

The following example illustrates how to load the StudFarm example from the Samples/oobn directory of the HUGIN installation and print the belief in node John.offspring:

Code: [Select]
ghapi = NET.addAssembly ('C:\Program Files\Hugin Expert\Hugin Researcher 7.4\HDE7.4CS\Lib\hugincs-7.4-2.0-x64.dll');
cc = HAPI.ClassCollection();
cc.ParseClasses ('C:\Program Files\Hugin Expert\Hugin Researcher 7.4\Samples\oobn\StudFarm.net', HAPI.DefaultClassParseListener);
c = cc.GetClassByName ('StudFarm');
d = c.CreateDomain ();
d.Compile ();
d.GetNodeByName ('John.offspring').GetBelief (0)
ans =

    0.2734

Notice that all classes of the example are stored as a class collection in a single file.

Pages: [1] 2 3 ... 6