Dear Colleagues, Probabilistic graphical models (PGMs) have become a popular statistical modelling tool with remarkable impact on disciplines like data mining and machine learning, because their most outstanding features are their clear semantics and interpretability. Bayesian inference methods naturally embed into PGMs, providing them with efficient and sound techniques for estimating both structure and parameters. Bayesian inference has been the key to the application of PGMs in specially demanding domains like streaming data analysis, where the models need to be frequently updated when new data arrives.
There are, however, a number of open issues concerning scalability, which is especially relevant in big data domains. In general, approximate techniques are employed, including variational inference and Markov Chain Monte Carlo. This Special Issue seeks original contributions covering aspects of Bayesian methods for learning PGMs from data and efficient algorithms for probabilistic inference in PGMs. Papers covering relevant modelling issues are also welcome, including papers dealing with data stream modelling, Bayesian change point detection, feature selection and automatic relevance determination. Even though entirely theoretical papers are within the scope of this Special Issue, contributions including a thorough experimental analysis of the methodological advances are particularly welcome, so that the impact of the proposed methods can be appropriately determined in terms of performance over benchmark datasets. Keywords Bayesian networks Probabilistic Graphical Models Bayesian methods Cross Entropy Methods Variational Inference Bayesian Data Stream Modelling Monte Carlo methods for PGMs Link: https://www.mdpi.com/journal/entropy/special_issues/graphical_models <https://www.mdpi.com/journal/entropy/special_issues/graphical_models> Deadline for paper submission: 1 March 2021 Prof. Rafael Rumí Prof. Antonio Salmerón Guest Editors — Antonio Salmerón Cerdán Department of Mathematics University of Almería http://www.ual.es/personal/asalmero
_______________________________________________ uai mailing list uai@engr.orst.edu https://it.engineering.oregonstate.edu/mailman/listinfo/uai
