Download the program booklet.

Conference Program at a Glance

Tuesday – Day 1 (Sep 6)
08:00 – 08:50	Registration
08:50 – 09:00	Opening
09:00 – 10:20	Session “Causality 1” (4 talks)
10:20 – 10:40	Coffee break
10:40 – 11:20	Session “Causality 2” (2 talks)
11:20 – 11:30	Break
11:30 – 12:30	Tutorial (Di Mauro and Vergari)
12:30 – 14:00	Conference lunch
14:00 – 15:00	Session “Missing Data” (3 talks)
15:00 – 15:20	Coffee break
15:20 – 16:20	Session “Sensitivity Analysis” (3 talks)
16:20 – 16:35	Break
16:35 – 17:55	Session “Representation” (4 talks)
18:30 – 20:00	Welcome reception

Wednesday – Day 2 (Sep 7)
09:00 – 10:20	Session “Classification” (4 talks)
10:20 – 10:40	Coffee break
10:40 – 11:20	Session “Relational Models” (2 talks)
11:20 – 11:30	Break
11:30 – 12:30	Invited talk (Darwiche)
12:30 – 14:00	Lunch
14:00 – 15:00	Invited talk (Consonni)
15:00 – 15:20	Coffee break
15:20 – 16:20	Session “Continuous Variables” (3 talks)
16:20 – 16:35	Break
16:35 – 17:35	Session “Estimation” (3 talks)
17:35 – 17:50	Break
17:50 – 18:20	General meeting
19:00 – 23:00	Conference dinner

Thursday – Day 3 (Sep 8)
09:00 – 10:20	Session “Structural Learning” (4 talks)
10:20 – 10:40	Coffee break
10:40 – 11:20	Session “Dynamic Models 1” (2 talks)
11:20 – 11:30	Break
11:30 – 12:30	Invited talk (Cozman)
12:30 – 14:00	Conference lunch
14:00 – 15:00	Session “Dynamic Models 2” (3 talks)
15:00 – 15:15	Coffee break (short)
15:15 – 16:15	Tutorial (de Campos and Kwisthout)
16:15 – 16:30	Break
16:30 – 17:30	Session “Complexity” (3 talks)
17:30 – 18:30	Round table (PGMs, software, and applications)

Friday – Day 4 (Sep 9)
09:00 – 10:20	Session “Inference” (4 talks)
10:20 – 10:40	Coffee break
10:40 – 11:20	Session “Sum-Product Nets” (2 talks)
11:20 – 11:30	Break
11:30 – 12:30	Invited (OR Days) talk (Svensson)
12:30 – 12:40	Closing

Session “Causality 1”

Chair: Ross Shachter

• Christiane Görgen and Jim Q. Smith.
  A differential approach to causality in staged trees.
• Daniel Malinsky and Peter Spirtes. Estimating causal effects with ancestral graph Markov models.
• Antti Hyttinen, Sergey Plis, Matti Järvisalo, Frederick Eberhardt, and David Danks. Causal discovery from subsampled time series data by constraint optmization.
• Juan Miguel Ogarrio, Peter Spirtes, and Joe Ramsey. A hybrid causal search algorithm for latent variable models (slides).

Session “Causality 2”

Chair: Pekka Parviainen

• Elena Sokolova, Martine Hoogman, Perry Groot, Tom Claassen, and Tom Heskes. Computing lower and upper bounds on the probability of causal statements.
• Jose Peña. Learning acyclic directed mixed graphs from observations and interventions (slides).

Session “Missing Data”

Chair: Marco Scutari

• Bence Bolgár and Péter Antal. Bayesian matrix factorization with non-random missing data using informative Gaussian process priors and soft evidences.
• Martin Plajner and Jiří Vomlel. Student skill models in adaptive testing.
• Konstantinos Sechidis, Matthew Sperrin, Emily Petherick, and Gavin Brown. Estimating mutual information in under-reported variables.

Session “Sensitivity Analysis”

Chair: Alessandro Antonucci

• Silja Renooij. Evidence evaluation: a study of likelihoods and independence.
• Jasper De Bock. Reintroducing credal networks under epistemic irrelevance.
• Janneke Bolt. Bayesian networks: a combined tuning heuristic.

Session “Representation”

Chair: Jiří Vomlel

• Yang Xiang and Qian Jiang. Compressing Bayes net CPTs with persistent leaky causes.
• Maxime Gasse and Alex Aussem. Identifying the irreducible disjoint factors of a multivariate probability distribution.
• Ross Shachter. Decisions and dependence in influence diagrams.
• Péter Marx, András Millinghoffer, Gabriella Juhász and Péter Antal. Joint Bayesian modelling of internal dependencies and revant multimorbidities of a heterogeneous disease.

Session “Classification”

Chair: Giorgio Corani

• Eugene Dementiev and Norman Fenton. Bayesian torrent classification by file name and size only.
• Nicola Di Mauro, Antonio Vergari, and Floriana Esposito. Multilabel classification with cutset networks.
• Yi Tan, Prakash Shenoy, Moses Chan, and Paul Romberg. On construction of hybrid logistic regression-naive Bayes model for classification.
• Marco Benjumeda, Concha Bielza, and Pedro Larrañaga. Learning tractable multidimensional Bayesian network classifiers.

Session “Relational Models”

Chair: Johan Kwisthout

• Nourhene Ettouzi, Philippe Leray, and Montassar Ben Messaoud. An exact approach to learning probabilistic relational model.
• Denis Mauá and Fabio Cozman. The effect of combination functions on the complexity of relational Bayesian networks.

Session “Continuous Variables”

Chair: Antonio Salmerón

• Kiran Karra and Lamine Milli. Hybrid copula Bayesian networks.
• Jidapa Kraisangka and Marek Druzdzel. Making large Cox’s proportional hazard models tractable in Bayesian networks.
• Evangelia Kyrimi and William Marsh. A progressive explanation of inference in “hybrid” Bayesian networks for supporting clinical decision making.

Session “Estimation”

Chair: Pedro Larrañaga

• Marco Cattaneo. Conditional probability estimation.
• Priya Krishnan Sundararajan and Ole Mengshoel. A genetic algorithm for learning parameters in Bayesian networks using expectation maximization.
• Eva Endres and Thomas Augustin. Statistical matching of discrete data by Bayesian networks.

Session “Structural Learning”

Chair: Cassio P. de Campos

• Marco Scutari. An empirical-Bayes score for discrete Bayesian networks.
• Milan Studený and James Cussens. The chordal graph polytope for learning decomposable models (slides).
• Eunice Yuh-Jie Chen, Arthur Choi, and Adnan Darwiche. On pruning with the MDL score.
• Pekka Parviainen and Samuel Kaski. Bayesian networks for variable groups.

Session “Dynamic Models 1”

Chair: Marek Druzdzel

• Carlos Morales and Serafín Moral. Regression methods applied to flight variables for situational awarenes estimation using dynamic Bayesian networks.
• Marcos Bueno, Arjen Hommersom, Peter Lucas, Sicco Verwer, and Alexis Linard. Learning complex uncertain states changes via asymmetric hidden Markov models: an industrial case.

Session “Dynamic Models 2”

Chair: Concha Bielza

• Thomas Geier, Michael Glodek, Georg Layher, Heiko Neumann, Susanne Biundo, and Günther Palm. On stacking probabilistic temporal models with bidirectional information flow.
• Manxia Liu, Arjen Hommersom, Maarten van der Heijden, and Peter Lucas. Learning parameters of hybrid time Bayesian networks.
• Marcus Bendtsen. Regime aware learning.

Session “Complexity Theory”

Chair: Milan Studený

• Fabio Cozman and Denis Mauá. Probabilistic graphical models specified by probabilistic logic programs: semantics and complexity.
• Cory Butz, André dos Santos, and Jhonatan Oliveira. Relevant path separation: a faster method for testing independencies in Bayesian networks.
• Johan Kwisthout. The parameterized complexity of approximate inference in Bayesian networks (slides).

Session “Inference”

Chair: Jose Peña

• Cory Butz, Jhonatan Oliveira, André dos Santos, and Anders Madsen. On Bayesian network inference with simple propagation.
• Andrés Masegosa, Ana Martínez, Helge Langseth, Thomas Nielsen, Antonio Salmerón, Darío Ramos-López, and Anders Madsen. d-VMP: distributed variational message passing.
• Ivar Simonsson and Petter Mostad. Exact inference on conditional linear Gamma-Gaussian Bayesian networks.
• Darí­o Ramos-López, Antonio Salmerón, Rafael Rumí, Ana Martínez, Thomas Nielsen, Andrés Masegosa, Helge Langseth, and Anders Madsen. Scalable MAP inference in Bayesian networks based on a Map-Reduce approach (slides).

Session “Sum-Product Networks”

Chair: Nicola di Mauro

• Priyank Jaini, Abdullah Rashwan, Han Zhao, Yue Liu, Ershad Banijamali, Zhitang Chen, and Pascal Poupart. Online algorithms for sum-product networks with continuous variables.
• Mazen Melibari, Pascal Poupart, Prashant Doshi, and George Trimponias. Dynamic sum product networks for tractable inference on sequence data.