Incremental Compilation of Object-Oriented Bayesian Networks
نویسنده
چکیده
Object-oriented paradigms have been applied to Bayesian networks to provide a modular structure which allows greater flexibility and robustness. These object-oriented Bayesian networks may be used over larger and more complex domains. However, as the networks get larger, the computational cost of triangulation and junction tree construction grows. The process of creating new junction trees when changes are made to object-oriented Bayesian networks is particular inefficient. Bangso et al [1] describe a method for using the networks maximal prime subgraph decomposition to only reconstruct the parts of the junction trees effected by these changes. This project is an implementation of that algorithm.
منابع مشابه
Incremental compilation of Bayesian networks
Most methods for exact probability propaga tion in Bayesian networks do not carry out the inference directly over the network, but over a secondary structure known as a junc tion tree or a join tree (JT). The process of obtaining a JT is usually termed compilation. As compilation is usually viewed as a whole process; each time the network is modified, a new compilation process has to be perfo...
متن کاملModular Compilation Strategies for Aspect-Oriented Constructs
In our previous work, we presented an aspect-oriented intermediate language, named Nu, to preserve design modularity in object code. Nu is based on two primitives: bind and remove. We showed that maintaining modularity in object code significantly improved the incremental compilation time of aspect-oriented programs. The key contribution of this work is a set of compilation strategies to Nu for...
متن کاملPreserving Separation of Concerns through Compilation A Position Paper
Today’s aspect-oriented programming (AOP) languages provide software engineers with new possibilities for keeping conceptual concerns separate at the source code level. For a number of reasons, current aspect weavers sacrifice this separation in transforming source to object code (and thus the very term weaving). In this paper, we argue that sacrificing modularity has significant costs, especia...
متن کاملObject-Oriented Dynamic Bayesian Network-Templates for Modelling Mechatronic Systems
The object-oriented paradigma is a new but proven technology for modelling mechatronics, i.e. multidisciplinary modelling. For many reasons the object-oriented approach is very much desirable also for qualitative models in system design, diagnosis or verification. Bayesian networks are a very robust technology for qualitative probabilistic modelling. In this paper we present a first approach in...
متن کامل