International Conference on Lightning Protection 

About: International Conference on Lightning Protection is an academic conference. The conference publishes majorly in the area(s): Lightning & Lightning strike. Over the lifetime, 2144 publications have been published by the conference receiving 25354 citations.

A Statistical Learning Method for Logic Programs with Distribution Semantics.

Abstract: 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 generalization of the traditional least model semantics, it can capture semantics of diverse information processing systems ranging from Bayesian networks to Hidden Markov models to Boltzmann machines in a single framework with mathematical rigor. Thus symbolic computation and statistical modeling are integrated at semantic level. With this new semantics, we propose a statistical learning schema based on the EM algorithm known in statistics. It enables logic programs to learn from examples, and to adapt to the surrounding environment. We implement the schema for a subclass of logic programs called BS-programs.

Abduction Compared with Negation by Failure.

Abstract: Horn clause logic programming can be extended to include abduction with integrity constraints. In the resulting extension of logic programming, negation by failure can be simulated by making negative conditions abducible and by imposing appropriate denials and disjunctions as integrity constraints. This gives an alternative semantics for negation by failure, which generalises the stable model semantics of negation by failure. The abductive extension of logic programming extends negation by failure in three ways: (1) computation can be perfonned in alternative minimal models, (2) positive as well as negative conditions can be made abducible, and (3) other integrity constraints can also be accommodated. * This paper was written while the first author was at Imperial College. 235 Introduction The tenn "abduction" was introduced by the philosopher Charles Peirce [1931] to refer to a particular kind of hypothetical reasoning. In the simplest case, it has the fonn: From A and A fB infer B as a possible "explanation" of A. Abduction has been given prominence in Charniak and McDennot's [1985] "Introduction to Artificial Intelligence", where it has been applied to expert systems and story comprehension. Independently, several authors have developed deductive techniques to drive the generation of abductive hypotheses. Cox and Pietrzykowski [1986] construct hypotheses from the "dead ends" of linear resolution proofs. Finger and Genesereth [1985] generate "deductive solutions to design problems" using the "residue" left behind in resolution proofs. Poole, Goebel and Aleliunas [1987] also use linear resolution to generate hypotheses. All impose the restriction that hypotheses should be consistent with the "knowledge base". Abduction is a fonn of non-monotonic reasoning, because hypotheses which are consistent with one state of a knowledge base may become inconSistent when new knowledge is added. Poole [1988] argues that abduction is preferable to noh-monotonic logics for default reasoning. In this view, defaults are hypotheses fonnulated within classical logic rather than conclusions derived withln some fonn of non-monotonic logic. The similarity between abduction and default reasoning was also pointed out in [Kowalski, 1979]. In this paper we show how abduction can be integrated with logic programming, and we concentrate on the use of abduction to generalise negation by failure. Conditional Answers Compared with Abduction In the simplest case, a logic program consists of a set of Horn Clauses, which are used backward to_reduce goals to sub goals. The initial goal is solved when there are no subgollls left;