The potential influence diagram is a generalization of the standard "conditional" influence diagram, a directed network representation for probabilistic inference and decision analysis [Ndilikilikesha, 1991). It allows efficient inference calculations corresponding exactly to those on undirected graphs. In this paper, we explore the relationship between potential and conditional influence diagr...

Interactive dynamic influence diagrams (I-DIDs) offer a transparent and semantically clear representation for the sequential decision-making problem over multiple time steps in the presence of other interacting agents. Solving I-DIDs exactly involves knowing the solutions of possible models of the other agents, which increase exponentially with the number of time steps. We present a method of s...

We consider the problem of learning about and comparing the consequences of dynamic treatment strategies on the basis of observational data. We formulate this within a probabilistic decision-theoretic framework. Our approach is compared with related work by Robins and others: in particular, we show how Robins’s ‘G-computation’ algorithm arises naturally from this decision-theoretic perspective....

We propose to exploit Decision Networks (DN) for the analysis of attack/defense scenarios. We show that DN extend both the modeling and the analysis capabilities of formalisms based on Attack Trees, which are the main reference model in such a context. Uncertainty can be addressed at every system level and a decisiontheoretic analysis of the risk and of the selection of the best countermeasures...

In building Bayesian belief networks, the elic­ itation of all probabilities required can be a major obstacle. We learned the extent of this often-cited observation in the construc­ tion of the probabilistic part of a complex influence diagram in the field of cancer treat­ ment. Based upon our negative experiences with existing methods, we designed a new method for probability elicitation from ...