Abstract
In order to handle real-world problems, state-of-the-art probabilistic logic and learning frameworks, such as ProbLog, reduce the expensive inference to an efficient Weighted Model Counting. To do so ProbLog employs a sequence of transformation steps, called an inference pipeline. Each step in the probabilistic inference pipeline is called a pipeline component. The choice of the mechanism to implement a component can be crucial to the performance of the system. In this paper we describe in detail different ProbLog pipelines. Then we perform a empirical analysis to determine which components have a crucial impact on the efficiency. Our results show that the Boolean formula conversion is the crucial component in an inference pipeline. Our main contributions are the thorough analysis of ProbLog inference pipelines and the introduction of new pipelines, one of which performs very well on our benchmarks.
Access provided by Autonomous University of Puebla. Download to read the full chapter text
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Bryant, R.E.: Graph-based algorithms for boolean function manipulation. IEEE Trans. Comput. 35(8), 677–691 (1986)
Chen, W., Swift, T., Warren, D.S.: Efficient top-down computation of queries under the well-founded semantics. J. Log. Program. 24(3), 161–199 (1995)
Darwiche, A.: A compiler for deterministic, decomposable negation normal form. In: Dechter, R., Sutton, R.S. (eds). AAAI/IAAI, pp. 627–634. AAAI Press/MIT Press (2002)
Darwiche, A.: New advances in compiling CNF into decomposable negation normal form. In: Proceedings of the 16th European Conference on Artificial Intelligence, pp. 328–332 (2004)
Darwiche, A.: Modeling and Reasoning with Bayesian Networks. Cambridge University Press (2009) (chapter 12)
Darwiche, A., Marquis, P.: A knowledge compilation map. Journal of Artificial Intelligence Research 17, 229–264 (2002)
De Raedt, L., Kimmig, A., Toivonen, H.: ProbLog: a probabilistic prolog and its application in link discovery. In Proceedings of the 20th International Joint Conference on Artificial Intelligence, pp. 2468–2473. AAAI Press (2007)
Fierens, D., Van Den Broek, G., Renkens, J., Shterionov, D., Gutmann, B., Thon, I., Janssens, G., de Raedt, L.: Inference and learning in probabilistic logic programs using weighted boolean formulas. Theory and Practice of Logic Programming, Special Issue on Probability, Logic and Learning 15(3), 358–401 (2015)
Fierens, D., Van den Broeck, G., Thon, I., Gutmann, B., De Raedt, L.: Inference in probabilistic logic programs using weighted CNF’s. In: Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, pp. 211–220 (2011)
Gutmann, B., Thon, I., De Raedt, L.: Learning the parameters of probabilistic logic programs from interpretations. In: Gunopulos, D., Hofmann, T., Malerba, D., Vazirgiannis, M. (eds.) ECML PKDD 2011, Part I. LNCS, vol. 6911, pp. 581–596. Springer, Heidelberg (2011)
Janhunen, T.: Representing normal programs with clauses. In: Proc. of the 16th European Conference on Artificial Intelligence, pp. 358–362. IOS Press (2004)
Kimmig, A., Demoen, B., De Raedt, L., Costa, V.S., Rocha, R.: On the implementation of the probabilistic logic programming language ProbLog. Theory and Practice of Logic Programming 11, 235–262 (2011)
Mantadelis, T.: Efficient Algorithms for Prolog Based Probabilistic Logic Programming. PhD thesis, Informatics Section, Department of Computer Science, Faculty of Engineering Science, November 2012. Janssens, Gerda (supervisor)
Mantadelis, T., Janssens, G.: Dedicated tabling for a probabilistic setting. In: Hermenegildo, M.V., Schaub, T. (eds) ICLP (Technical Communications), vol. 7 of LIPIcs, pp. 124–133. Schloss Dagstuhl - Leibniz-Zentrum fuer Informatik (2010)
Muise, C., McIlraith, S.A., Beck, J.C., Hsu, E.I.: Dsharp: fast d-DNNF compilation with sharpSAT. In: Kosseim, L., Inkpen, D. (eds.) Canadian AI 2012. LNCS, vol. 7310, pp. 356–361. Springer, Heidelberg (2012)
De Raedt, L., Kersting, K.: Probabilistic inductive logic programming. In: De Raedt, L., Frasconi, P., Kersting, K., Muggleton, S.H. (eds.) Probabilistic Inductive Logic Programming. LNCS (LNAI), vol. 4911, pp. 1–27. Springer, Heidelberg (2008)
Rauzy, A., Châtelet, E., Dutuit, Y., Bérenguer, C.: A practical comparison of methods to assess sum-of-products. Rel. Eng. & Sys. Safety 79(1), 33–42 (2003)
Shterionov, D., Janssens, G.: Data acquisition and modeling for learning and reasoning in probabilistic logic environment. In: Antunes, L., Pinto, H.S., Prada, R., Trigo, P. (eds) Proceedings of the 15th Portuguese Conference on Artificial Intelligence, pp. 298–312 (2011)
Shterionov, D., Janssens, G.: Crucial components in probabilistic inference pipelines: Data and results. Technical report, KU Leuven, 2014. Ref. number CW679. http://www.cs.kuleuven.be/publicaties/rapporten/cw/CW679.pdf
Shterionov, D., Renkens, J., Vlasselaer, J., Kimmig, A., Meert, W., Janssens, G.: The most probable explanation for probabilistic logic programs with annotated disjunctions. In: Proceedings of the 24th International Conference on Inductive Logic Programming
Vennekens, J., Verbaeten, S., Bruynooghe, M.: Logic programs with annotated disjunctions. In: Demoen, B., Lifschitz, V. (eds.) ICLP 2004. LNCS, vol. 3132, pp. 431–445. Springer, Heidelberg (2004)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2015 Springer International Publishing Switzerland
About this paper
Cite this paper
Shterionov, D., Janssens, G. (2015). Implementation and Performance of Probabilistic Inference Pipelines. In: Pontelli, E., Son, T. (eds) Practical Aspects of Declarative Languages. PADL 2015. Lecture Notes in Computer Science(), vol 9131. Springer, Cham. https://doi.org/10.1007/978-3-319-19686-2_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-19686-2_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-19685-5
Online ISBN: 978-3-319-19686-2
eBook Packages: Computer ScienceComputer Science (R0)