Results 1 
8 of
8
A comparison of graphical techniques for decision analysis
 European Journal of Operational Research
, 1994
"... Abstract: Recently, we proposed a new method for representing and solving decision problems based on the framework of valuationbased systems. The new representation is called a valuation network, and the new solution method is called a fusion algorithm. In this paper, we compare valuation networks ..."
Abstract

Cited by 25 (11 self)
 Add to MetaCart
(Show Context)
Abstract: Recently, we proposed a new method for representing and solving decision problems based on the framework of valuationbased systems. The new representation is called a valuation network, and the new solution method is called a fusion algorithm. In this paper, we compare valuation networks to decision trees and influence diagrams. We also compare the fusion algorithm to the backward recursion method of decision trees and to the arcreversal method of influence diagrams.
Independency relationships and learning algorithms for singly connected networks
, 1998
"... Graphical structures such as Bayesian networks or M arkov networks are very useful tools for representing irrelevance or independency relationships, and they may be used to efficiently perform reasoning tasks. Singly connected networks are important specific cases where there is no more than one un ..."
Abstract

Cited by 19 (10 self)
 Add to MetaCart
Graphical structures such as Bayesian networks or M arkov networks are very useful tools for representing irrelevance or independency relationships, and they may be used to efficiently perform reasoning tasks. Singly connected networks are important specific cases where there is no more than one undirected path connecting each pair of variables. The aim of this paper is to investigate the kind of properties that a dependency model must verify in order to be equivalent to a singly connected graph structure, as a way of driving automated discovery and construction of singly connected networks in data. The main results are the characterizations of those dependency models which are isomorphic to singly connected graphs (either via the dseparation criterion for directed acyclic graphs or via the separation criterion for undirected graphs), as well as the development of efficient algorithms for learning singly connected graph representations of dependency models.
An Algorithm for Finding Minimum dSeparating Sets in Belief Networks
 Proceedings of the twelfth Conference of Uncertainty in Artificial Intelligence
, 1996
"... The criterion commonly used in directed acyclic graphs (dags) for testing graphical independence is the wellknown dseparation criterion. It allows us to build graphical representations of dependency models (usually probabilistic dependency models) in the form of belief networks, which make possibl ..."
Abstract

Cited by 17 (4 self)
 Add to MetaCart
(Show Context)
The criterion commonly used in directed acyclic graphs (dags) for testing graphical independence is the wellknown dseparation criterion. It allows us to build graphical representations of dependency models (usually probabilistic dependency models) in the form of belief networks, which make possible an easy interpretation and management of independence relationships, without reference to numerical parameters (conditional probabilities). In this paper we study the following combinatorial problem: to find the minimum dseparating set for two nodes in a dag. This set would represent the minimum information necessary to prevent these two nodes to influence each other. The solution of this basic problem and of some of its extensions can be useful in several ways, as we will see later. Our solution is based on a twosteps process: first, we reduce the original problem to the simpler one of finding a minimum separating set in an undirected graph, and second, we develop an algorithm for solvi...
Characterizations of Decomposable Dependency Models
 Journal of Artificial Intelligence Research
, 1996
"... Decomposable dependency models possess a number of interesting and useful properties. This paper presents new characterizations of decomposable models in terms of independence relationships, which are obtained by adding a single axiom to the wellknown set characterizing dependency models that are i ..."
Abstract

Cited by 4 (2 self)
 Add to MetaCart
(Show Context)
Decomposable dependency models possess a number of interesting and useful properties. This paper presents new characterizations of decomposable models in terms of independence relationships, which are obtained by adding a single axiom to the wellknown set characterizing dependency models that are isomorphic to undirected graphs. We also briefly discuss a potential application of our results to the problem of learning graphical models from data. 1. Introduction Graphical models are knowledge representation tools commonly used by an increasing number of researchers, particularly from the Artificial Intelligence and Statistics communities. The reason for the success of graphical models is their capacity to represent and handle independence relationships, which have proved crucial for the efficient management and storage of information (Pearl, 1988). There are different kinds of graphical models, although we are particularly interested in undirected and directed graphs (which, in a proba...
Guest Editorial New perspectives on Causal Networks: the ®rst CaNew workshop
, 1998
"... www.elsevier.com/locate/ijar We are pleased to introduce a selection of the papers presented at the 1998 workshop on `Causal Networks from Inference to Data Mining', CaNew '98, [59]. This workshop was initiated from the feeling, shared by the organizers and cochairs, that the ®eld of Baye ..."
Abstract
 Add to MetaCart
www.elsevier.com/locate/ijar We are pleased to introduce a selection of the papers presented at the 1998 workshop on `Causal Networks from Inference to Data Mining', CaNew '98, [59]. This workshop was initiated from the feeling, shared by the organizers and cochairs, that the ®eld of Bayesian and, in general, Causal Networks deserved special attention from the international research community. We had a growing feeling that several areas had been neglected in research or deserved more attention. The common background of the editors and cochairs being in Machine Learning, we felt that some ideas that had been long been in use in Machine Learning had not been applied to Causal Networks. However, we also felt that other aspects dealing with the knowledge representation aspects of the Causal Network formalism were also of interest, namely, the construction of networks that used di€erent uncertainty formalisms, new inference methods and the relationship between the classical interpretation of Causal Network and the new ones. The rest of the Workshop Programme Committee members had a similar feeling about that and we tried to convey this by introducing in the workshop title both ends of the Causal Networks research spectrum: from inference to Data Mining. We comment in more detail in Section 3 the opportunities that, from our point of view, lay hidden between both.
Research Note Characterizations of Decomposable Dependency Models
"... Decomposable dependency models possess a numberofinteresting and useful properties. This paper presents new characterizations of decomposable models in terms of independence relationships, which are obtained by adding a single axiom to the wellknown set characterizing dependency models that are iso ..."
Abstract
 Add to MetaCart
(Show Context)
Decomposable dependency models possess a numberofinteresting and useful properties. This paper presents new characterizations of decomposable models in terms of independence relationships, which are obtained by adding a single axiom to the wellknown set characterizing dependency models that are isomorphic to undirected graphs. We also brie y discuss a potential application of our results to the problem of learning graphical models from data. 1.
Independency Relationships in Singly Connected Networks
, 1994
"... Graphical structures such as causal networks or Markov networks are very useful tools for representing irrelevance or independency relationships. Singly connected networks are important specific cases where there is no more than one undirected path connecting each pair of variables. The aim of this ..."
Abstract
 Add to MetaCart
(Show Context)
Graphical structures such as causal networks or Markov networks are very useful tools for representing irrelevance or independency relationships. Singly connected networks are important specific cases where there is no more than one undirected path connecting each pair of variables. The aim of this paper is to investigate the kind of properties that a dependency model must verify in order to be equivalent to a singly connected graph structure, either via the dseparation criterion for directed acyclic graphs or via the separation criterion for undirected graphs. The main results are the characterizations of those dependency models which are isomorphic to singly connected graphs, as well as the development of efficient algorithms for learning singly connected graph representations of dependency models. keywords: Independency relations, graphical models, singly connected networks, learning algorithms. 1 Introduction Graphs have become common knowledge representation tools capable of e...