Evidence-invariant Sensitivity Bounds
This addresses the computational challenge for practitioners using probabilistic networks in real-life scenarios, but it is incremental as it builds on existing sensitivity analysis methods.
The paper tackles the problem of sensitivity analysis in probabilistic networks being dependent on evidence, which makes complete analysis infeasible due to many possible observation combinations. It presents a method for studying evidence-invariant sensitivities by establishing bounds on parameter variations that do not change the most likely value of a variable of interest.
The sensitivities revealed by a sensitivity analysis of a probabilistic network typically depend on the entered evidence. For a real-life network therefore, the analysis is performed a number of times, with different evidence. Although efficient algorithms for sensitivity analysis exist, a complete analysis is often infeasible because of the large range of possible combinations of observations. In this paper we present a method for studying sensitivities that are invariant to the evidence entered. Our method builds upon the idea of establishing bounds between which a parameter can be varied without ever inducing a change in the most likely value of a variable of interest.