ProbLog#Implementations

A probabilistic fact is a pair $(p,\; a)$ with $a$ a ground atom and $p\; \backslash in\; [0,\; 1]$ the probability of $a$ being true. A rule is defined by an atom $h$, called the head, and a finite set of $n$ literals $\backslash \{\; b\_1,\; b\_2,\; .\; .\; .,\; b\_n\; \backslash \}$, called the body.

ProbLog programs consist of a set of probabilistic facts $\backslash mathcal\{F\}$ and a set of rules $\backslash mathcal\{R\}$. Using the distribution semantics, a probability distribution is defined over the two-valued well-founded models of the atoms in the program. The probability of a model is defined as $$P(M)\; =\; \backslash prod\_\{l\; \backslash in\; M\}\; P(l)$$ where the product runs over all the literals in the model $M$. For a query atom $q$ the distribution semantics defines a probability for the query

$$P(q)\; =\; \backslash sum\_\{M\; \backslash models\; q\}\; P(M)\; =\; \backslash sum\_\{M\; \backslash models\; q\}\; \backslash prod\_\{l\; \backslash in\; M\}\; P(l)$$

in which the sum runs over all the models where $q$ is true.

ProbLog supports multiple tasks:

• Probabilistic inference: calculate $P(q)$
• Most probable explanation: calculate the most probable model probability $\backslash max\_\{M\; \backslash models\; q\}\; P(M)$
• Sampling: generate samples of $q$
• Learning from interpretations: learn the probabilities of ProbLog programs from data

ProbLog can for example be used to calculate the probability of getting wet given the probabilities for rain and the probabilities that someone brings an umbrella as follows:

0.4 :: rain(weekday).

0.9 :: rain(weekend).

0.8 :: umbrella_if_rainy(Day).

0.2 :: umbrella_if_dry(Day).

umbrella(Day) :- rain(Day), umbrella_if_rainy(Day).

umbrella(Day) :- \+rain(Day), umbrella_if_dry(Day).

wet(Day) :- rain(Day), \+umbrella(Day).

query(\+wet(weekend)).

The last rule before the query states that someone gets wet if it rains and no umbrella was brought. When ProbLog is asked to solve the "probabilistic inference" task, the query asks for the probability to stay dry on a weekend day. When solving the "most probable explanation" task, ProbLog will return the most likely reason for staying dry, i.e. because it is not raining or because the person has an umbrella.

The ProbLog language has been implemented as a YAP Prolog library (ProbLog 1). and as a stand-alone Python framework (ProbLog 2)

The source code of ProbLog 2 is licensed under Apache License, Version 2.0 and available on GitHub. The ProbLog language has also been implemented as part of the cplint probabilistic logic programming package for SWI-Prolog, YAP and XSB.{{Cite web |title=cplint – AI@UNIFE |url=https://ai.unife.it/cplint/ |access-date=2023-11-13 |language=en-US}}

ProbLog has been extended or used as inspiration for several different variants, including:

• DeepProbLog extends ProbLog by allowing the probability to be parametrized by a neural network.
• DTProblog extends ProbLog with decision theory. The utility of a strategy is defined as the expected reward for its execution in the presence of probabilistic effects.
• DC-ProbLog extends ProbLog with distributional facts, meaning that instead of probabilities, a logic atom has a corresponding continuous probability distribution instead.
• aProbLog generalizes ProbLog by allowing any commutative semiring instead of just probabilities.
• ProbFOIL: given a set of ProbLog facts as a probabilistic relational database, ProbFOIL finds a set of probabilistic rules to predict the facts of one relation based on all other relations.

• PRISM: Programming in statistical modeling
• ICL: Independent Choice Logic
• CP-Logic: Language of causal probabilistic events
• LPAD: Logic programs with annotated disjunctions
• Distributional clauses: A probabilistic logic language for hybrid relational domains

• ProbLog homepage
• ProbLog docs
• ProbLog repository

{{Reflist|30em|refs=

{{cite web|url=http://dtai.cs.kuleuven.be/problog|title=ProbLog: Probabilistic Programming|website=dtai.cs.kuleuven.be}}

{{cite web|url=https://problog.readthedocs.io/en/latest/|title=ProbLog: ProbLog 2.1 documentation|website=problog.readthedocs.io}}

{{cite web|url=https://github.com/ML-KULeuven/problog|title=ProbLog GitHub repository|website=github.com|date=12 October 2022 }}

{{cite web|url=https://dtai.cs.kuleuven.be/problog/problog1/installation.html|title=ProbLog1|website=dtai.cs.kuleuven.be}}

{{cite web|url=https://dtai.cs.kuleuven.be/ml/systems/DC/|title=Distributional Clauses|website=dtai.cs.kuleuven.be}}

{{cite web|url=https://ml.unife.it/pita/|title=PITA: Probabilistic Inference with Tabling and Answer subsumption|website=ml.unife.it}}

{{cite web|url=https://rjida.meijo-u.ac.jp/prism/|title=PRISM: PRogramming In Statistical Modeling |website=rjida.meijo-u.ac.jp}}

{{cite conference

|last1 = De Raedt|first1 = Luc|last2 = Kimmig|first2 = Angelika|last3 = Toivonen|first3 = Hannu|title = ProbLog: A Probabilistic Prolog and Its Application in Link Discovery|conference = IJCAI|date = November 2007|volume = 7}}

{{cite journal

|last1 = De Raedt|first1 = Luc|last2 = Kimmig|first2 = Angelika|title = Probabilistic (logic) programming concepts|journal = Machine Learning|date = 2015|volume = 100|number = 1|pages = 5–47| doi=10.1007/s10994-015-5494-z | s2cid=3166992 |doi-access = free}}

{{cite conference

|last1 = Van den Broeck|first1 = Guy|last2 = Thon|first2 = Ingo|last3 = Van Otterlo|first3 = Martijn|last4=De Raedt|first4=Luc|title = DTProbLog: A decision-theoretic probabilistic Prolog|book-title = Proceedings of the AAAI Conference on Artificial Intelligence|date = 2010|volume = 24|number = 1}}

{{cite conference|last1 = Manhaeve|first1 = Robin|last2 = Dumancic|first2 = Sebastijan|last3 = Kimmig|first3 = Angelika|last4 = Demeester|first4 = Thomas|last5 = De Raedt|first5 = Luc|title = DeepProbLog: Neural Probabilistic Logic Programming|conference = NeurIPS 2018, Thirty-second Conference on Neural Information Processing Systems|date = 2018|pages = 3753–3760}}

{{cite conference|last1 = Fierens|first1 = D|last2 = Van den Broeck|first2 = G.|last3 = Bruynooghe |first3 = M.|last4 = De Raedt|first4 = L.|title = Constraints for probabilistic logic programming|conference = Proceedings of the NIPS Probabilistic Programming Workshop|date = 2012|pages = 1–4}}

{{cite conference|last1 = Kimmig|first1 = A.|last2 = Van den Broeck|first2 = G.|last3 = De Raedt|first3 = L.|title = An algebraic Prolog for reasoning about possible worlds|conference = Proceedings of the Twenty-Fifth AAAI Conference on Artificial Intelligence|date = 2011|pages = 209–214}}

{{cite conference|last1 = Vennekens|first1 = Joost|last2 = Denecker|first2 = Marc|last3 = Bruynooghe|first3 = Maurice |title = CP-logic: A language of causal probabilistic events and its relation to logic programming|conference = Theory and practice of logic programming|date = 2009|pages = 245–308|volume = 9|arxiv = 0904.1672}}

{{cite book|last1 = Poole|first1 = David|chapter = The independent choice logic and beyond|title = Probabilistic Inductive Logic Programming| series=Lecture Notes in Computer Science |date = 2008|pages = 222–243|volume = 4911| editor1= Luc Raedt|editor2= Paolo Frasconi|editor3= Kristian Kersting|editor4= Stephen Muggleton |publisher=Springer |doi=10.1007/978-3-540-78652-8_8 | isbn=978-3-540-78651-1 }}

}}

Category:Probabilistic software

Category:Programming paradigms

Category:Nondeterministic programming languages

Category:Computational statistics

Category:Python (programming language) scientific libraries

Category:Logic programming languages