Zooming in on trade-offs in qualitative probabilistic networks (original) (raw)

Enhanced qualitative probabilistic networks for resolving trade-offs

Artificial Intelligence, 2008

Qualitative probabilistic networks were designed to overcome, to at least some extent, the quantification problem known to probabilistic networks. Qualitative networks abstract from the numerical probabilities of their quantitative counterparts by using signs to summarise the probabilistic influences between their variables. One of the major drawbacks of these qualitative abstractions, however, is the coarse level of representation detail that does not provide for indicating strengths of influences. As a result, the trade-offs modelled in a network often remain unresolved upon inference. We present an enhanced formalism of qualitative probabilistic networks to provide for a finer level of representation detail. An enhanced qualitative probabilistic network differs from a basic qualitative network in that it distinguishes between strong and weak influences. Now, if a strong influence is combined, upon inference, with a conflicting weak influence, the sign of the net influence may be readily determined. Enhanced qualitative networks are purely qualitative in nature, as basic qualitative networks are, yet allow for resolving more trade-offs upon inference.

Refining reasoning in qualitative probabilistic networks

1995

Abstract In recent years there has been a spate of papers describing systems for probabilisitic reasoning which do not use numerical probabilities. In some cases the simple set of values used by these systems make it impossible to predict how a probability will change or which hypothesis is most likely given certain evidence. This paper concentrates on such situations, and suggests a number of ways in which they may be resolved by re ning the representation.

Efficient Reasoning in Qualitative Probabilistic Networks

National Conference on Artificial Intelligence, 1993

Qualitative Probabilistic Networks (QPNs) are anabstraction of Bayesian belief networks replacingnumerical relations by qualitative influences andsynergies [ Wellman, 1990b ] . To reason in a QPNis to find the effect of new evidence on each node interms of the sign of the change in belief (increaseor decrease). We introduce a polynomial time algorithmfor reasoning in QPNs, based on localsign propagation.

From qualitative to quantitative probabilistic networks

Quantifi cation is well known to be a major ob stacle in the construction of a probabilistic net work, especially when relying on human experts for this purpose. The construction of a qualitative probabilistic network has been proposed as an initial step in a network's quantifi cation, since the qualitative network can be used to gain prelimi nary insight in the projected network's reasoning behaviour. We extend on this idea and present a new type of network in which both signs and numbers are specifi ed; we further present an associated algorithm for probabilistic inference. Building upon these semi-qualitative networks, a probabilistic network can be quantified and stud ied in a stepwise manner. As a result, modelling inadequacies can be detected and amended at an early stage in the quantifi cation process.

Belief propagation in Qualitative Probabilistic Networks

1993

Qualitative probabilistic networks (QPNs) are an abstraction of inuence diagrams and Bayesian belief networks replacing numerical relations by qualitative inuences and synergies. To reason in a QPN is to nd the eect of decision or new evidence on a variable of interest in terms of the sign of the change in belief (increase or decrease). We review our work on qualitative belief propagation, a computationally ecient reasoning scheme based on local sign propagation in QPNs. Qualitative belief propagation, unlike the existing graph-reduction algorithm, preserves the network structure and determines the eect of evidence on all nodes in the network. We show how this supports meta-level reasoning about the model and automatic generation of intuitive explanations of probabilistic reasoning.

Inference in qualitative probabilistic networks revisited

International Journal of Approximate Reasoning, 2009

Qualitative probabilistic networks (QPNs) are basically qualitative derivations of Bayesian belief networks. Originally, QPNs were designed to improve the speed of the construction and calculation of these networks, at the cost of specificity of the result. The formalism can also be used to facilitate cognitive mapping by means of inference in sign-based causal diagrams. Whatever the type of application, any computer based use of QPNs requires an algorithm capable of propagating information throughout the networks. Such an algorithm was developed in the 1990s. This polynomial time sign-propagation algorithm is explicitly or implicitly used in most existing QPN studies.

Building probabilistic networks: "Where do the numbers come from?" - Guest editors' introduction

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2000

the relationships between them from domain experts is comparable, to at least some extent, to knowledge engineering for other artificial-intelligence representations and, although it may require significant effort, is generally considered doable. The last task in building a probabilistic network is to obtain the probabilities that are required for its quantitative part. This task often appears more daunting: "Where do the numbers come from?" is a commonly asked question. The three tasks in building a probabilistic network are, in principle, performed one after the other. Building a network, however, often requires a careful trade-off between the desire for a large and rich model to obtain accurate results on the one hand, and the costs of construction and maintenance and the complexity of probabilistic inference on the other hand. In practice, therefore, building a probabilistic network is a process that iterates over these tasks until a network results that is deemed requisite.

Introducing situational signs in qualitative probabilistic networks

International Journal of Approximate Reasoning, 2005

A qualitative probabilistic network is a graphical model of the probabilistic influences among a set of statistical variables in which each influence is associated with a qualitative sign. A non-monotonic influence between two variables is associated with the ambiguous sign ' ?', which indicates that the actual sign of the influence depends on the state of the network. The presence of such ambiguous signs is undesirable as it tends to lead to uninformative results upon inference. In this paper, we argue that, in each specific state of the network, the effect of a non-monotonic influence is unambiguous. To capture the current effect of the influence, we introduce the concept of situational sign. We show how situational signs can be used upon inference and how they are updated as the state of the network changes. By means of a real-life qualitative network in oncology, we show that the use of situational signs can effectively forestall uninformative results upon inference.

Context-specific sign-propagation in qualitative probabilistic networks

2002

Qualitative probabilistic networks are qualitative abstractions of probabilistic networks, summarising probabilistic influences by qualitative signs. As qualitative networks model influences at the level of variables, knowledge about probabilistic influences that hold only for specific values cannot be expressed. The results computed from a qualitative network, as a consequence, can be weaker than strictly necessary and may in fact be rather uninformative.

Qualitative Probabilistic Networks for Planning Under Uncertainty

Machine Intelligence and Pattern Recognition, 1988

Bayesian networks provide a probabilistic seman tics for qualitative assertions about likelihood. A qualitative reasoner based on an algebra over these assertions can derive further conclusions about the influence of actions. While the conclusions are much weaker than those computed from complete probability distributions, they are still valuable for suggesting potential actions, eliminating obviously inferior plans, identifying important tradeoffs, and explaining probabilistic models.