Tsai-Ching Lu

Most relevant explanation in Bayesian networks

A major inference task in Bayesian networks is explaining why some variables are observed in their particular states using a set of target variables. Existing methods for solving this problem often generate explanations that are either too simple (underspecified) or too complex (overspecified). In this paper, we introduce a method called Most Relevant Explanation (MRE) which finds a partial instantiation of the target variables that maximizes the generalized Bayes factor (GBF) as the best explanation for the given evidence. Our study shows that GBF has several theoretical properties that enable MRE to automatically identify the most relevant target variables in forming its explanation. In particular, conditional Bayes factor (CBF), defined as the GBF of a new explanation conditioned on an existing explanation, provides a soft measure on the degree of relevance of the variables in the new explanation in explaining the evidence given the existing explanation. As a result, MRE is able to automatically prune less relevant variables from its explanation. We also show that CBF is able to capture well the explaining-away phenomenon that is often represented in Bayesian networks. Moreover, we define two dominance relations between the candidate solutions and use the relations to generalize MRE to find a set of top explanations that is both diverse and representative. Case studies on several benchmark diagnostic Bayesian networks show that MRE is often able to find explanatory hypotheses that are not only precise but also concise.

Interactive construction of graphical decision models based on causal mechanisms

We propose a framework for building graphical decision models from individual causal mechanisms. Our approach is based on the work of Simon [Simon, H.A., 1953. Causal ordering and identifiability. In: Hood, W.C., Koopmans, T.C. (Eds.), Studies in Econometric Method. Cowles Commission for Research in Economics. Monograph No. 14. John Wiley and Sons Inc., New York, NY, pp. 49-74 (Ch. III)], who proposed a causal ordering algorithm for explicating causal asymmetries among variables in a self-contained set of structural equations. We extend the causal ordering algorithm to under-constrained sets of structural equations, common during the process of problem structuring. We demonstrate that the causal ordering explicated by our extension is an intermediate representation of a modelers understanding of a problem and that the process of model construction consists of assembling mechanisms into self-contained causal models. We describe ImaGeNIe, an interactive modeling tool that supports mechanism-based model construction and demonstrate empirically that it can effectively assist users in constructing graphical decision models.

Causal Models, Value of Intervention, and Search for Opportunities

While algorithms for influence diagrams allow for computing the optimal setting for decision variables, they offer no guidance in generation of decision variables, arguably a critical stage of decision making. A decision maker confronted with a complex system may not know which variables to best manipulate to achieve a desired objective. We introduce the concept of search for opportunities which amounts to identifying the set of decision variables and computing their optimal settings, given an objective expressed by a utility function. Search for opportunities is built on value of intervention in causal models.