Sensitivity Analysis for Probability Assessments in Bayesian Networks
This work addresses the challenge of improving model accuracy in expert systems for knowledge engineers, though it appears incremental as it builds on existing sensitivity analysis techniques.
The paper tackles the problem of aligning Bayesian network models with expert intuition by developing a methodology for analytically computing sensitivity values to measure the impact of parameter changes on target probabilities, which can guide knowledge elicitation and optimize parameter estimation using gradient descent.
When eliciting probability models from experts, knowledge engineers may compare the results of the model with expert judgment on test scenarios, then adjust model parameters to bring the behavior of the model more in line with the expert's intuition. This paper presents a methodology for analytic computation of sensitivity values to measure the impact of small changes in a network parameter on a target probability value or distribution. These values can be used to guide knowledge elicitation. They can also be used in a gradient descent algorithm to estimate parameter values that maximize a measure of goodness-of-fit to both local and holistic probability assessments.