Results 1  10
of
26
Welldefinedness and efficient inference for probabilistic logic programming under the distribution semantics. Theory and Practice of Logic Programming
 BIBLIOGRAPHY
"... ar ..."
Program Updating by Incremental and Answer Subsumption Tabling
"... Abstract. We explore the use of stateoftheart Logic Programming (LP) implementation techniques to exploit their use in addressing a classical nonmonotonic reasoning problem, that of LP program updates, with incidence on representing change, i.e. internal or self and external or world changes. We ..."
Abstract

Cited by 7 (7 self)
 Add to MetaCart
(Show Context)
Abstract. We explore the use of stateoftheart Logic Programming (LP) implementation techniques to exploit their use in addressing a classical nonmonotonic reasoning problem, that of LP program updates, with incidence on representing change, i.e. internal or self and external or world changes. We do so starting from a given LP update language and a given LP implementation system. We propose and foster a novel conceptual approach to program updates implementation that especially exploits two recent features of tabling in logic programming (in XSB Prolog): incremental and answer subsumption tabling. Our approach, termed EVOLP/R, is based on and follows the paradigm and constructs of Evolving Logic Programs (EVOLP), but simplifies it at first by restricting updates to fluents only. Rule updates are nevertheless achieved via the mechanism of rule name fluents, placed in rules ’ bodies, permitting to turn rules on or off, through assertions or retractions of their corresponding unique name fluents. Incremental tabling of fluents allows to automatically maintain – at engine level – the consistency of program states, analogously to an assumption based truthmaintenance
Tabled abduction in logic programs
 Technical Communication of ICLP 2013). Theory and Practice of Logic Programming, Online Supplement
, 2013
"... Abstract. Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. On the other hand, current Prolog systems, with their tabling mechanisms, are mature enough to facilitate ..."
Abstract

Cited by 6 (6 self)
 Add to MetaCart
(Show Context)
Abstract. Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. On the other hand, current Prolog systems, with their tabling mechanisms, are mature enough to facilitate the introduction of tabling abductive solutions (tabled abduction) into them. Our contributions are as follows: (1) We conceptualize tabled abduction for abductive normal logic programs, permitting abductive solutions to be reused, from one abductive context to another. The approach relies on a transformation into tabled logic programs that makes use of the dual transformation, and enables efficiently handling the problem of abduction under negative goals, by introducing dual positive counterparts for them. (2) We realize tabled abduction in TABDUAL, a system implemented in XSB Prolog, allowing dualization byneed only. (3) We refine the dual transformation in the context of TABDUAL to permit executing programs with variables and nonground queries. (4) We foster pragmatic approaches in TABDUAL to cater to all varieties of loops in normal logic programs, now complicated by abduction. (5) We evaluate TABDUAL in practice by examining five variants, according to various evaluation objectives. (6) We detail how TABDUAL can be applied to declarative debugging and decision making. (7) Finally, we refer to related work, and discuss TABDUAL’s correctness, complexity, and features that could migrate to the engine level, in Logic Programming systems wanting to encompass tabled abduction.
Implementing tabled abduction in logic programs.” Submitted to Doctoral Symposium on Artificial Intelligence (SDIA), Available at http://centria.di.fct.unl. pt/~lmp/publications/onlinepapers/implementing_tabdual.pdf
, 2013
"... Abstract. Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. If they become tabled, then abductive solutions can be reused, even from one abductive context to another. ..."
Abstract

Cited by 4 (4 self)
 Add to MetaCart
(Show Context)
Abstract. Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. If they become tabled, then abductive solutions can be reused, even from one abductive context to another. On the other hand, current Prolog systems, with their tabling mechanisms, are mature enough to facilitate the introduction of tabling abductive solutions (tabled abduction) into them. We recently proposed and published a conception of tabled abduction with its prototype, TABDUAL, implemented in XSBProlog. We detail here subsequent progress that has been made on the implementation aspect of TABDUAL, towards its more practical use.
TABDUAL: a tabled abduction system for logic programs. Accepted in IfCoLog Journal of Logics and their Applications. Available from http://goo.gl/lcQGes
, 2015
"... Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. On the other hand, current Prolog systems, with their tabling mechanisms, are mature enough to facilitate the intro ..."
Abstract

Cited by 3 (3 self)
 Add to MetaCart
(Show Context)
Abduction has been on the back burner in logic programming, as abduction can be too difficult to implement, and costly to perform, in particular if abductive solutions are not tabled. On the other hand, current Prolog systems, with their tabling mechanisms, are mature enough to facilitate the introduction of tabling abductive solutions (tabled abduction) into them. Our contributions are as follows. First, we conceptualize a tabled abduction technique for abductive normal logic programs, permitting abductive solutions to be reused, from one abductive context to another. The approach is underpinned by the theory of ABDUAL and relies on a transformation into tabled logic programs. It particularly makes use of the dual transformation of ABDUAL that enables efficiently handling the problem of abduction under negative goals, by introducing dual positive counterparts for them. Second, we realize this tabled abduction technique in TABDUAL, a system implemented in XSB Prolog. The implementation poses several challenges to concretely realize the abstract theory of ABDUAL, e.g., by taking care of all varieties of loops
Structure learning of probabilistic logic programs by searching the clause space
 CoRR/arXiv:1309.2080
, 2013
"... ar ..."
Optimizing Inference for Probabilistic Logic Programs Exploiting Independence and Exclusiveness
 Italian Conference on Computational Logic
"... Abstract. Probabilistic Logic Programming (PLP) is gaining popularity due to its many applications in particular in Machine Learning. An important problem in PLP is how to compute the probability of queries. PITA is an algorithm for solving such a problem that exploits tabling, answer subsumption an ..."
Abstract

Cited by 2 (2 self)
 Add to MetaCart
(Show Context)
Abstract. Probabilistic Logic Programming (PLP) is gaining popularity due to its many applications in particular in Machine Learning. An important problem in PLP is how to compute the probability of queries. PITA is an algorithm for solving such a problem that exploits tabling, answer subsumption and Binary Decision Diagrams (BDDs). PITA does not impose any restriction on the programs. Other algorithms, such as PRISM, achieve a higher speed by imposing two restrictions on the program, namely that subgoals are independent and that clause bodies are mutually exclusive. Another assumption that simplifies inference is that clause bodies are independent. In this paper we present the algorithms PITA(IND,IND) and PITA(OPT). PITA(IND,IND) assumes that subgoals and clause bodies are independent. PITA(OPT) instead first checks whether these assumptions hold for subprograms and subgoals: if they hold, PITA(OPT) uses a simplified calculation, otherwise it resorts to BDDs. Experiments on a number of benchmark datasets show that PITA(IND,IND) is the fastest on datasets respecting the assumptions while PITA(OPT) is a good option when nothing is known about a dataset. 1
Speeding Up Inference for Probabilistic Logic Programs
"... Probabilistic Logic Programming (PLP) allows to represent domains containing many entities connected by uncertain relations and has many applications in particular in Machine Learning. PITA is a PLP algorithm for computing the probability of queries that exploits tabling, answer subsumption and Bina ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
Probabilistic Logic Programming (PLP) allows to represent domains containing many entities connected by uncertain relations and has many applications in particular in Machine Learning. PITA is a PLP algorithm for computing the probability of queries that exploits tabling, answer subsumption and Binary Decision Diagrams (BDDs). PITA does not impose any restriction on the programs. Other algorithms, such as PRISM, reduce computation time by imposing restrictions on the program, namely that subgoals are independent and that clause bodies are mutually exclusive. Another assumption that simplifies inference is that clause bodies are independent. In this paper we present the algorithms PITA(IND,IND) and PITA(OPT). PITA(IND,IND) assumes that subgoals and clause bodies are independent. PITA(OPT) instead first checks whether these assumptions hold for subprograms and subgoals: if they do, PITA(OPT) uses a simplified calculation, otherwise it resorts to BDDs. Experiments on a number of benchmark datasets show that PITA(IND,IND) is the fastest on datasets respecting the assumptions while PITA(OPT) is a good option when nothing is known about a dataset.
Viterbi training in PRISM
 In Proceedings of the ICML12 Workshop on Statistical Relational Learning
, 2012
"... VT (Viterbi training), or hard EM, is an efficient way of parameter learning for probabilistic models with hidden variables. Given an observation y, it searches for a state of hidden variables x that maximizes p(x, y  θ) by coordinate ascent on parameters θ. In this paper we introduce VT to PRISM, ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
VT (Viterbi training), or hard EM, is an efficient way of parameter learning for probabilistic models with hidden variables. Given an observation y, it searches for a state of hidden variables x that maximizes p(x, y  θ) by coordinate ascent on parameters θ. In this paper we introduce VT to PRISM, a logicbased probabilistic modeling system for generative models. VT improves PRISM’s probabilistic modeling in two ways. First although generative models are said to be inappropriate for discrimination tasks in general, when parameters are learned by VT, models often show good discrimination performance. We conducted two parsing experiments with probabilistic grammars while learning parameters by a variety of inference methods, i.e. VT,EM,MAP and VB. The result is that VT achieves the best parsing accuracy among them in both experiments. Second since VT always deals with a single probability of a single explanation, Viterbi explanation, the exclusiveness condition imposed on PRISM programs is no more required when we learn parameters by VT. PRISM with VT thus allows us to write inclusive clause bodies, learn parameters and compute Viterbi explanations. 1.
Probabilistic Ontologies in Datalog+/
"... Abstract. In logic programming the distribution semantics is one of the most popular approaches for dealing with uncertain information. In this paper we apply the distribution semantics to the Datalog+/ language that is grounded in logic programming and allows tractable ontology querying. In the re ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
(Show Context)
Abstract. In logic programming the distribution semantics is one of the most popular approaches for dealing with uncertain information. In this paper we apply the distribution semantics to the Datalog+/ language that is grounded in logic programming and allows tractable ontology querying. In the resulting semantics, called DISPONTE, formulas of a probabilistic ontology can be annotated with an epistemic or a statistical probability. The epistemic probability represents a degree of confidence in the formula, while the statistical probability considers the populations to which the formula is applied. The probability of a query is defined in terms of finite set of finite explanations for the query. We also compare the DISPONTE approach for Datalog+/ ontologies with that of Probabilistic Datalog+/ where an ontology is composed of a Datalog+/theory whose formulas are associated to an assignment of values for the random variables of a companion Markov Logic Network. 1