Abstract
ParaGnosis (https://doi.org/10.5281/zenodo.7312034, https://zenodo.org/badge/latestdoi/560170574, Alternative url: https://github.com/gisodal/paragnosis, Demo url: https://github.com/gisodal/paragnosis/blob/main/DEMO.md) is an open-source tool that supports inference queries on Bayesian networks through weighted model counting. In the knowledge compilation step, the input Bayesian network is encoded as propositional logic and then compiled into a knowledge base in decision diagram representation. The tool supports various diagram formats, including the Weighted-Positive Binary Decision Diagram (WPBDD) which can concisely represent discrete probability distributions.
Once compiled, the probabilistic knowledge base can be queried in the inference step. To efficiently implement both steps, ParaGnosis uses simulated annealing to split the knowledge base into a number of partitions. This further reduces the decision diagram size and crucially enables parallelism in both the compilation and the inference steps. Experiments demonstrate that this partitioned approach, in combination with the WPBDD representation, can outperform other approaches in the knowledge compilation step, at the cost of slightly more expensive inference queries. Additionally, the tool can attain 15-fold parallel speedups using 64 cores.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
CUDD is available at http://vlsi.colorado.edu/~fabio/.
- 3.
The SDD compiler is available at http://reasoning.cs.ucla.edu/sdd/.
- 4.
The WPBDD compiler is available at https://github.com/gisodal/paragnosis/.
- 5.
Ace is available at http://reasoning.cs.ucla.edu/ace/.
- 6.
Dlib is available at http://dlib.net/.
- 7.
HUGIN is available at http://www.hugin.com/.
References
Bahar, R.I., et al.: Algebraic decision diagrams and their applications. In: Proceedings of 1993 International Conference on Computer Aided Design, pp. 188–191 (1993)
Bartocci, E., Kovács, L., Stankovič, M.: Mora - automatic generation of moment-based invariants. In: TACAS 2020. LNCS, vol. 12078, pp. 492–498. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-45190-5_28
Bryant, R.E.: Graph-based algorithms for Boolean function manipulation. Trans. Comput. 100, 677–691 (1986)
Ceylan, I.I., Darwiche, A., Van den Broeck, G.: Open-world probabilistic databases: semantics, algorithms, complexity. Artif. Intell. 295, 103–123 (2021)
Chavira, M., Darwiche, A.: On probabilistic inference by weighted model counting. Artif. Intell. 172, 772–799 (2008)
Chen, Y., Darwiche, A.: On the definition and computation of causal treewidth. In: The 38th Conference on Uncertainty in Artificial Intelligence (2022)
Choi, Y.K., Santillana, C., Shen, Y., Darwiche, A., Cong, J.: FPGA acceleration of Probabilistic Sentential Decision Diagrams with high-level synthesis. ACM Trans. Reconfigurable Technol. Syst. 16(2), 1–22 (2022)
Clarke, E.M., Fujita, M., McGeer, P.C., McMillan, K., Yang, J.C.-Y., Zhao, X.: Multi-terminal binary decision diagrams: An efficient data structure for matrix representation (2001)
Dal, G.H., Kosters, W.A., Takes, F.W.: Fast diameter computation of large sparse graphs using GPUs. In: International Conference on Parallel, Distributed and Network-Based Processing, pp. 632–639 (2014)
Dal, G.H., Laarman, A.W., Hommersom, A., Lucas, P.J.: A compositional approach to probabilistic knowledge compilation. Int. J. Approximate Reason. 138, 38–66 (2021)
Dal, G.H., Laarman, A.W., Lucas, P.J.F.: Parallel probabilistic inference by weighted model counting. In: International Conference on Probabilistic Graphical Models, pp. 97–108 (2018)
Dal, G.H., Lucas, P.J.F.: Weighted positive binary decision diagrams for exact probabilistic inference. Int. J. Approx. Reason. 90, 411–432 (2017)
Dal, G.H., Michels, S., Lucas, P.J.F.: Reducing the cost of probabilistic knowledge compilation. Mach. Learn. Res. 73 141–152 (2017)
Darwiche, A.: SDD: A new canonical representation of propositional knowledge bases. In: International Joint Conference on Artificial Intelligence, vol. 22, pp. 819 (2011)
Darwiche, A., Hirth, A.: On the (complete) reasons behind decisions. J. Logic, Lang. Inform. Ages 18, 1–26 (2022)
Darwiche, A., Marquis, P.: A knowledge compilation map. J. Artif. Intell. Res. 17, 229–264 (2002)
Darwiche, A., Marquis, P.: On quantifying literals in Boolean logic and its applications to explainable AI. J. Artif. Intell. Res. 72, 285–328 (2021)
van Dijk, T., Laarman, A., van de Pol, J.: Multi-core BDD operations for symbolic reachability. Electron. Notes Theor. Comput. Sci. 296, 127–143 (2013)
van Dijk, T., van de Pol, J.: Sylvan: multi-core framework for decision diagrams. Int. J. Softw. Tools Technol. Transfer 19, 675–696 (2017)
Dudek, J., Phan, V., Vardi, M.: ADDMC: Weighted model counting with algebraic decision diagrams. In: International Conference on Artificial Intelligence, pp. 1468–1476 (2020)
Hillmich, S., Burgholzer, L., Stogmuller, F., Wille, R.: Reordering decision diagrams for quantum computing is harder than you might think. In: International Conference on Reversible Computation, pp. 93–107 (2022)
Hitzler, P., Sarker, M.K.: Tractable Boolean and arithmetic circuits. Neuro-Symbolic Artif. Intell.: State Art 342, 146 (2022)
Madsen, A.L., Lang, M., Kjaerulff, U.B., Jensen, F.: The Hugin tool for learning Bayesian networks. In: European Conference on Symbolic and Quantitative Approaches to Reasoning and Uncertainty, pp. 594–605 (2003)
Miller, D.M., Thornton, M.A.: QMDD: A decision diagram structure for reversible and quantum circuits. In: 36th International Symposium on Multiple-Valued Logic, pp. 30–36 (2006)
Narayan, A., Jain, J., Fujita, M., Sangiovanni-Vincentelli, A.: Partitioned ROBDDs-a compact, canonical and efficiently manipulable representation for Boolean functions. In: Proceedings of International Conference on Computer Aided Design, pp. 547–554 (1996)
Judea Pearl. Bayesian networks (2011)
Peham, T., Burgholzer, L., Wille, R.: Equivalence checking paradigms in quantum circuit design: A case study. In: Proceedings of the 59th Design Automation Conference, pp. 517–522 (2022)
Sahoo, D.: A partitioning methodology for BDD-based verification. In: International Conference on Formal Methods in Computer-Aided Design, pp. 399–413 (2004)
Sanner, S., McAllester, D.: Affine algebraic decision diagrams (AADDs) and their application to structured probabilistic inference. In: International Joint Conference on Artificial Intelligence 2005, pp. 1384–1390 (2005)
Shih, A., Choi, A., Darwiche, A.: Compiling Bayesian network classifiers into decision graphs. In: International Conference on Artificial Intelligence, vol. 33, pp. 7966–7974 (2019)
Sokukcu, M., Sakar, C.: Risk analysis of collision accidents during underway sts berthing maneuver through integrating fault tree analysis (FTA) into Bayesian network. Appl. Ocean Res. 126, 103–124 (2022)
Tafertshofer, P., Pedram, M.: Factored edge-valued binary decision diagrams. Formal Methods Syst. Design 10(2), 243–270 (1997)
Viamontes, G.F., Markov, I.L., Hayes, J.P.: Improving gate-level simulation of quantum circuits. Quant. Inform. Process., 2, 347–380 (2003)
Wilson, N.: Decision diagrams for the computation of semiring valuations. In: Proceedings of the 19th International Joint Conference on Artificial Intelligence, pp. 331–336 (2005)
Zulehner, A., Hillmich, S., Wille, R.: How to efficiently handle complex values? Implementing decision diagrams for quantum computing. In: 2019 International Conference on Computer-Aided Design, pp. 1–7 (2019)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Dal, G.H., Laarman, A., Lucas, P.J.F. (2023). ParaGnosis: A Tool for Parallel Knowledge Compilation. In: Caltais, G., Schilling, C. (eds) Model Checking Software. SPIN 2023. Lecture Notes in Computer Science, vol 13872. Springer, Cham. https://doi.org/10.1007/978-3-031-32157-3_2
Download citation
DOI: https://doi.org/10.1007/978-3-031-32157-3_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-32156-6
Online ISBN: 978-3-031-32157-3
eBook Packages: Computer ScienceComputer Science (R0)