Probabilistic logic programs are logic programs in which some of the facts are annotated with probabilities.

ProbLog is a tool that allows you to intuitively build programs that do not only encode complex interactions between a large sets of heterogenous components but also the inherent uncertainties that are present in real-life situations.

The engine tackles several tasks such as computing the marginals given evidence and learning from (partial) interpretations. ProbLog is a suite of efficient algorithms for various inference tasks. It is based on a conversion of the program and the queries and evidence to a weighted Boolean formula. This allows us to reduce the inference tasks to well-studied tasks such as weighted model counting, which can be solved using state-of-the-art methods known from the graphical model and knowledge compilation literature.

ProbLog is a Python package and can be embedded in Python or Java. Its knowledge base can be represented as Prolog/Datalog facts, CSV-files, SQLite database tables, through functions implemented in the host environment or combinations hereof.

The Language. Probabilistic Logic Programming

ProbLog makes it easy to express complex, probabilistic models.

0.3::stress(X) :- person(X).
0.2::influences(X,Y) :- person(X), person(Y).

smokes(X) :- stress(X).
smokes(X) :- friend(X,Y), influences(Y,X), smokes(Y).

0.4::asthma(X) :- smokes(X).



Get to know the language using our interactive tutorial, check out our publications or try the above example directly in our online editor.

For an introduction, please consult the following papers

  • Inference and learning in probabilistic logic programs using weighted Boolean formulas, Daan Fierens, Guy Van den Broeck, Joris Renkens, Dimitar Shterionov, Bernd Gutmann, Ingo Thon, Gerda Janssens, and Luc De Raedt. Theory and Practice of Logic Programming, 2015. PDF
  • ProbLog: A probabilistic Prolog and its application in link discovery, L. De Raedt, A. Kimmig, and H. Toivonen, Proceedings of the 20th International Joint Conference on Artificial Intelligence (IJCAI-07), Hyderabad, India, pages 2462-2467, 2007. PDF

The Engine. Powerful inference.

ProbLog2 is our second generation engine to reason with the ProbLog language. The current engine builds on logic programming, knowledge compilation, the distribution semantics and probabilistic, graphical models. It allows you to:

  • Compute marginal probabilities of any number of ground atoms in the presence of evidence.
  • Learn the parameters of the ProbLog program from partial interpretations.
  • Sample from a ProbLog program.
  • Solve decision theoretic problems

Online interface. No hassle.

You can try out ProbLog using our online editor.

Try tutorial

Try inference and learning

Download. Use, change, improve.

ProbLog2 binaries and source are available for download. The only requirement is Python 2.7 or 3. ProbLog2 is licensed under the Apache Public License 2.0.

ProbLog2 is available from the following sources:

Extended documentation is available on ReadTheDocs.

ProbLog2 can optionally make use of: SDDs, c2d and DSHARP. The previous version of ProbLog (ProbLog1) is available as part of YAP Prolog or can be downloaded from the old Problog1 webpages.

There is also an experimental variant of ProbLog that supports continuous variables: Distributional Clauses.

Need help? Fire any question or report a bug to our team.

Mailing list

There is a general ProbLog mailing list which you can use for all your ProbLog1 and ProbLog2 questions or bug reports. You can also consult the mailing list archive to see if your question has been asked before.

The ProbLog team.

ProbLog2 was developed in the DTAI group of KULeuven, by the following people (in alphabetical order).

Other people who contributed to ProbLog1 and 2 (in alphabetical order):