Probabilistic Networks for Automated Reasoning

2. A new type of causal models was formulated, based on embedding structural considerations in the language of sequential, temporal situation calculus. By using situation calculus as a basic language, we leverage its power to express complex, dynamically changing situations and, by relying on structural considerations, we can formulate an e ective theory of counterfactuals within the situationcalculus. We have shown that this hybrid approach can handle predictions, interventions, and counterfactuals using many of same techniques derived from the structural model approach.