First European Workshop on 

Probabilistic Graphical Models - PGM'02

Workshop Programme

 

Download in postscript

 

Tuesday 5

17:30 Bus from Madrid-Barajas to Cuenca.

21:00 Welcome Cocktail.

 

Wednesday 6

9:00 Registration

10:00 Opening

10:45 Coffee break

11:10 Session A: Propagation

13:50 Lunch

16:00 Special session: Tools presentation

17:30 Coffee break

18:00 Session B: Machine learning (statistical issues)

 

Thursday 7

9:00 Session C: Foundations

11:00 Coffee break

11:30 Session D: Influence diagrams

13:50 Lunch

16:00 Session E: Applications

17:30 Coffee break

18:00 Museum Visit.

21:00 Banquet

 

Friday 8

9:30 Session F: Learning Bayesian networks

11:00 Coffee break

11:30 Session G: Learning in Graphical Models

13:00 Closing

13:50 Lunch

 

11:10-13:40 Session A. Propagation

• Optimal Time-Space Tradeoff in Probabilistic Inference
  David Allen, Adnan Darwiche

• Factorisation of probability trees and its application to inference in Bayesian networks
  Irene Martínez, Serafín Moral, Carmelo Rodríguez, Antonio Salmerón

• Inferentially efficient propagation in non-decomposable Bayesian network with Hierarchical junction trees
  Roberto Puch, Jim Q. Smith, Concha Bielza

• Cooperative Verification of Agent Interface
  Yang Xiang, X. Chen

• A practical relaxation of constant-factor treewidth approximation algorithms
  Mark Hopkins, Adnan Darwiche

 

16:00-17:30 Special Session: Tools presentation

Note: This session has been sponsorised by the Spanish Association for Artificial Intelligence (AEPIA)

• HUGIN - The Tool for Bayesian Networks and Influence Diagrams
  Fran Jensen, Uffe B. Kjaerulff, Michael Lang, Anders L. Madsen
  

• Elvira: An Environment for Creating and Using Probabilistic Graphical Models.
  Serafín Moral (Elvira Consortium)

• Genie:
  Marek J. Druzdzel

18:00-19:00 Session B. Machine learning (statistical issues)

• Scaled Conjugate Gradients for Maximum Likelihood: An Empirical Comparison with the EM Algorithm
  Kristian Kersting, Niels Landwehr

• Estimating mixtures of truncated exponentials from data
  Serafín Moral, Rafael Rumí, Antonio Salmerón

 

9:00-11:00 Session C. Foundations

• Multicausal prior families, randomisation and essential graphs
  Alireza Daneshkhah, Jim. Q. Smith

• New Structures for Conditional Probability Tables
  Juan A. Fernández. del Pozo, Concha Bieza

• Logical Hidden Markov Models
  Kristian Kersting, T. Raiko, L. De Raedt

• Characterization of essential graphs by means of an operation of legal component merging
  Milan Studený

 

11:30-13:30 Session D: Influence Diagrams

• Probabilistic Decision Graphs - Combining verification and AI techniques for probabilistic inference
  Manfred Jaeger

• Causal Models, Value of Intervention, and Search for Opportunities
  Tsai-Ching Lu, Marek J. Druzdzel

• Approximate Solutions of Complex Influence Diagrams through MCMC methods.
  Miguel A. Virto, J. Martín, D. Ríos-Insua, A. Moreno

• An extension of lazy evaluation for influence diagrams avoiding redundant variables in the potentials
  Marta Vomlelová, Finn V. Jensen

 

16:00-17:30 Session E: Applications

• Reducing propagation effort in large polytrees: An application to information retrieval
  Luis de Campos, Juan M. Fernández, Juan Huete

• Continuous Speech Recognition Using Dynamic Bayesian Networks. A Fast Decoding Algorithm
  Murat Deviren, Khalid Daoudi

• Bayesian Networks in Educational Testing
  Jirí Vomlel

 

9:30-11:00 Session F: Learning Bayesian networks

• Restricted Bayesian Network Structure Learning
  Peter Lucas

• Floating Search Methods in Learning Bayesian Networks
  Rosa Blanco, Iñaki Inza, Pedro Larrañaga

• Unsupervised Learning of Bayesian Networks Via Estimation of Distribution Algorithms
  José M. Peña, José A. Lozano, Pedro Larrañaga

 

11:30 - 13:00 Session G: Learning in graphical models

• Graphical Models to causal discovery from data
  Luis de Campos, José A. Gámez, José M. Puerta

• Handling of deterministic relationships in constraint-based causal discovery
  Yusuf K. Yilmaz, Ethem Alpaydin, Levent Akin, Taner Bilgiç

• Learning Essential Graph Markov Models From Data
  Robert Castelo, Michael D. Perlman

 

Welcome Cocktail. Tuesday. 21:00

The Welcome Cocktail will be in the hall of the Science Museum of Castilla-La Mancha.
Address: Plaza de la Merced, 1. Cuenca.

 

Lunch (Wednesday, Thursday, Friday). 13:50

The lunch will be served at Restaurante Horno de las Campanas. This is the restaurant of Hotel Leonor de Aquitania
Address: C./ San Pedro, 60. Cuenca.

 

MCCM Visit. Thursday. 18:00

There would be a free guided tour to the Science Museum of Castilla-La Mancha.

 

 
Banquet. Thursday. 21:00

The Banquet/Conference dinner will be served at the Parador Nacional de Cuenca
Address: Paseo Hoz del Huecar, s/n. Cuenca.