Exploring Unknown Universes in Probabilistic Relational Models

Large probabilistic models are often shaped by a pool of known individuals (a universe) and relations between them. Lifted inference algorithms handle sets of known individuals for tractable inference. Universes may not always be known, though, or may only described by assumptions such as "small universes are more likely". Without a universe, inference is no longer possible for lifted algorithms, losing their advantage of tractable inference. The aim of this paper is to define a semantics for models with unknown universes decoupled from a specific constraint language to enable lifted and thereby, tractable inference.

[1]  Umut A. Acar,et al.  Adaptive inference on general graphical models , 2008, UAI.

[2]  David Poole,et al.  First-order probabilistic inference , 2003, IJCAI.

[3]  Guy Van den Broeck,et al.  Open World Probabilistic Databases (Extended Abstract) , 2016, Description Logics.

[4]  Luc De Raedt,et al.  Lifted Probabilistic Inference by First-Order Knowledge Compilation , 2011, IJCAI.

[5]  Guy Van den Broeck,et al.  Tractability through Exchangeability: A New Perspective on Efficient Probabilistic Inference , 2014, AAAI.

[7]  Stuart J. Russell,et al.  BLOG: Probabilistic Models with Unknown Objects , 2005, IJCAI.

[8]  Kristian Kersting,et al.  Exploiting symmetries for scaling loopy belief propagation and relational training , 2013, Machine Learning.

[9]  Stuart J. Russell,et al.  First-Order Open-Universe POMDPs , 2014, UAI.

[10]  Jesse Davis,et al.  Lifted Variable Elimination: Decoupling the Operators from the Constraint Language , 2013, J. Artif. Intell. Res..

[11]  Tanya Braun,et al.  Adaptive Inference on Probabilistic Relational Models , 2018, Australasian Conference on Artificial Intelligence.

[12]  Ronald Fagin,et al.  Combining Fuzzy Information from Multiple Systems , 1999, J. Comput. Syst. Sci..

[13]  Danna Zhou,et al.  d. , 1934, Microbial pathogenesis.

[14]  Seyed Mehran Kazemi,et al.  Knowledge Compilation for Lifted Probabilistic Inference: Compiling to a Low-Level Language , 2016, KR.

[15]  John B. Shoven,et al.  I , Edinburgh Medical and Surgical Journal.

[16]  Luc De Raedt,et al.  ProbLog: A Probabilistic Prolog and its Application in Link Discovery , 2007, IJCAI.

[17]  Norbert Fuhr,et al.  Probabilistic Datalog—a logic for powerful retrieval methods , 1995, SIGIR '95.

[18]  Matthew Richardson,et al.  Markov logic networks , 2006, Machine Learning.

[19]  Miroslaw Truszczynski,et al.  Answer set programming at a glance , 2011, Commun. ACM.

[20]  Tanya Braun,et al.  Preventing Groundings and Handling Evidence in the Lifted Junction Tree Algorithm , 2017, KI.