Gregory M. Provan

Learn More
We present several techniques for knowledge engineering of large belief networks (BNs) based on the our experiences with a network derived from a large medical knowledge base. The noisyMAX, a generalization of the noisy-OR gate, is used to model causal independence in a BN with multivalued variables. We describe the use of leak probabilities to enforce the(More)
Bayesian belief networks are being increasingly used as a knowledge representation for reasoning under uncertainty. Some researchers have questioned the practicality of obtaining the numerical probabilities with sucient precision to create belief networks for large-scale applications. In this work, we i n v estigate how precise the probabilities need to be(More)
A model-based diagnosis problem occurs when an observation is inconsistent with the assumption that the diagnosed system is not faulty. The task of a diagnosis engine is to compute diagnoses, which are assumptions on the health of components in the diagnosed system that explain the observation. In this paper, we extend Reiter’s well-known theory of(More)
We describe a new paradigm for implement­ ing inference in belief networks, which con­ sists of two steps: (1) compiling a belief network into an arithmetic expression called a Query DAG (Q-DAG); and (2) answering queries using a simple evaluation algorithm. Each non-leaf node of a Q-DAG represents a numeric operation, a number, or a symbol for evidence.(More)
Computing diagnoses in domains with continuously changing data is a difficult, but essential aspect of solving many problems. To address this task, this paper describes a dynamic influence diagram (ID) construction and updating system, DYNASTY, and its application to constructing a decision-theoretic model to diagnose acute abdominal pain, a domain in which(More)
Dempster-Shafer (DS) theory is formulated in terms of propositional logic, using the implicit notion of provability underlying DS theory. Dempster-Shafer theory can be modeled in terms of propositional logic by the tuple (~, p), where S is a set of propositional clauses and p is an assignment of mass to each clause Ei c ~. It is shown that the disjunction(More)
We propose a StochAstic Fault diagnosis AlgoRIthm, called Safari, which trades off guarantees of computing minimal diagnoses for computational efficiency. We empirically demonstrate, using the 74XXX and ISCAS85 suites of benchmark combinatorial circuits, that Safari achieves several orders-of-magnitude speedup over two well-known deterministic algorithms,(More)
This paper analyzes the theoretical under­ pinnings of recent proposals for comput ing Dempster-Shafer Belief functions f rom A T M S labels. Such proposals are intended to be a means of integrat ing symbolic and numeric rep­ resentation methods and of focusing search in the A T M S . This synthesis is formalized us­ ing graph theory, thus showing the relat(More)