Lorentz Center - PGM2004 - Probabilistic Graphical Models 2004 from 4 Oct 2004 through 8 Oct 2004
  Current Workshop  |   Overview   Back  |   Home   |   Search   |     

    PGM2004 - Probabilistic Graphical Models 2004
    from 4 Oct 2004 through 8 Oct 2004

Preliminary Program

Preliminary Program

Working sessions are meant for collaborative research and proposal writing, e.g. a proposal on a PGM Network to be submitted to the European Science Foundation.


Monday, 4th October, 2004


• 10:00–12:00 Registration (this includes allocation of offices)

• 10:00–11:00 Coffee & Tea in the Common Room

• 11:00–13:00 Parellel Working Sessions I

• 13:00–15:00 Lunch (Het Witte Huis)

• 15:00–16:00 Parellel Working Sessions II

• 16:00–16:15 Welcome to PGM’04: Peter Lucas

• 16:15–17:00 Invited Talk (Title to be announced): Prakash Shenoy

• 17:00–18:00 Wine & Cheese Party

• 19:30– Dinner (Het Witte Huis)


Tuesday, 5th October, 2004


• 9:30–11:00: Learning (Chair: Pedro Larranaga)


V. Romero, R. Rumi and A. Salmerón: Structural learning of Bayesian networks with mixtures of truncated exponentials

C. Riggelsen: Learning Bayesian network parameters from incomplete data using importance sampling

J. Roure Alcobé: An approximated MDL score for learning Bayesian networks


• 11:00–11:30 Coffee Break


• 11:30–13:00 Influence Diagrams and Decision Networks (Chair: Javier Díez)


A. Cano, M. Gómez and S. Moral: A forward-backward Monte Carlo method for solving influence diagrams


D. Garcia-Sanchez and M.J. Druzdzel: An efficient sampling algorithm for influence diagrams


J.A. Fernández del Pozo and C. Bielza: Sensitivity analysis and explanation of optimal decisions


• 13:30–15:00 Lunch (Het Witte Huis)


• 15:30–16:30 Probabilistic Reasoning (Chair: Marek Druzdzel)


D. Allen, A. Darwiche and J.D. Park: A greedy algorithm for time-space tradeoff in probabilistic inference


X. An, Y. Xiang and N. Cercone: Cooperative computation of Markov boundaries for efficient observation in multiagent probabilistic inference


• 16:30–17:00 Coffee Break


• 17:00–18:00 Probabilistic Reasoning (Chair: Kristian Olesen)


M. Arias and F.J. Díez: Operating with potentials of discrete variables


J.H. Bolt and L.C. van der Gaag: Decisiveness in loopy propagation


• 19:30– Dinner (Het Witte Huis)


Wednesday, 6th October, 2004


• 9:30–11:00 Learning (Chair: Peter Lucas)


C. Cotta and J. Muruzábal: On the learning of Bayesian network graph structures via evolutionary programming


A. Feelders and L.C. van der Gaag: Learning Bayesian network parameters under order constraints


G. Karčiauskas, T. Kočka, F.V. Jensen, P. Larraňaga and J.A. Lozano: Learning of latent class models by splitting and merging components


• 11:00–11:30 Coffee Break


• 11:30–13:00 Influence Diagrams and Decision Networks (Chair: Concha Bielza)


M. Luque and F.J. Díez: Variable elimination for influence diagrams with supervalue nodes


S.H. Nielsen, T.D. Nielsen and F.V. Jensen: Multi-currency influence diagrams


F.V. Jensen, T.D. Nielsen and P.P. Shenoy: Sequential influence diagrams: a unified asymmetry framework


• 13:30–15:00 Lunch (Het Witte Huis)


• 15:30–17:00 Applications of PGM (Chair: José Antonio Gamez)


M. van Gerven and P.J.F. Lucas: Using background knowledge to construct Bayesian classifiers for data-poor domains


J.M. Peňa: Learning and validating Bayesian network models of genetic regulatory networks


A. Cano, J.G. Castellano, A.R. Masegosa and S. Moral: Application of a

selective Gaussian naive Bayes model for di_use large-B-cell lymphoma classification


19:30– Dinner (Het Witte Huis)


Thursday, 7th October, 2004


• 9:30–11:00 Foundations (Chair: Milan Studenỳ)


L.E. Sucar: Parallel MDPs


R. Jiroušek: Decomposition of Bayesian networks induced by compositional models


L. Figueroa-Quiroz and J.Q. Smith: Causal Bayesian networks of time series

whose processes obey balance


• 11:00–11:30 Coffee Break


• 11:30–13:00 Probabilistic Reasoning (Chair: Linda van der Gaag)


B.R. Cobb and P.P. Shenoy: Inference in hybrid Bayesian networks with deterministic variables


Y. Xiang and X. Chen: Inference in multiply sectioned Bayesian networks with lazy propagation and linked junction forests


C. Yuan and M.J. Druzdzel: A comparison of the e_ectiveness of two heuristics

for importance sampling


• 13:30–15:00 Pancakes (Pannenkoeken) (Centre of Leiden)


• 15:00–21:00 Visit of museum “De Lakenhal” (15:00–17:00), site seeing of the centre of Leiden (17:00–19:30) and dinner at restaurant “Het Prentenkabinet” (19:30)


Friday, 8th October, 2004


• 10:00–11:00 Foundations (Chair: Antonio Salmerón)


M. Jaeger, J.D. Nielsen and T. Silander: Learning probabilistic decision graphs


M. Studeny and J. Vomlel: Transition between graphical and algebraic representatives of Bayesian network models


• 11:00–11:30 Coffee Break


• 11:30–12:30 Probabilistic Reasoning (Chair: Silja Renooij)


M. Chavira, A. Darwiche and M. Jaeger: Compiling relational Bayesian networks for exact inference


W. Wiegerinck: Approximate explanation of reasoning in Bayesian networks


• 13:30–15:00 Lunch (Het Witte Huis)


• 15:30–17:00 Parallel Working Sessions III