Parameter Learning of Logic Programs for Symbolic-Statistical Modeling

Parameter Learning of Logic Programs for Symbolic-Statistical Modeling

We propose a logical/mathematical framework for statistical parameter learning of parameterized logic programs, i.e. de nite clause programs containing probabilistic facts with a parameterized distribution. It extends the traditional least Herbrand model semantics in logic programming to distribution semantics , possible world semantics with a probability distribution which is unconditionally a...

A Graphical Method for Parameter Learning of Symbolic-Statistical Models

A Sparse Parameter Learning Method for Probabilistic Logic Programs

We propose a new parameter learning algorithm for ProbLog, which is an extension of a logic program that can perform probabilistic inferences. Our algorithm differs from previous parameter learning algorithms for probabilistic logic program (PLP) models on the point that it tries to reduce the number of probabilistic parameters contained in the estimated program. Since the amount of computation...

A Statistical Learning Method for Logic Programs with Distribution Semantics

When a joint distribution P F is given to a set F of facts in a logic program DB = F [R where R is a set of rules, we can further extend it to a joint distribution P DB over the set of possible least models of DB. We then de ne the semantics of DB with the associated distribution P F as P DB , and call it distribution semantics. While the distribution semantics is a straightforward generalizati...

PRISM: A Language for Symbolic-Statistical Modeling

We present an overview of symbolic-statistical modeling language PRISM whose programs are not only a probabil istic extension of logic programs but also able to learn f rom examples w i th the help of the EM learning algori thm. As a knowledge representation language appropriate for probabil istic reasoning, it can describe various types of symbolic-statistical modeling formalism known but unre...