Underlying semantics for IDS worlds
Erik J. Sandewall · 1995
Abstract The definition of IDS worlds given in Chapter 1 was quite broad. More specific classes of worlds can be obtained in two ways: first by imposing various constraints according to the ontological characteristics, and then by specifying exactly the effects of actions using action laws. Conversely, from the point of view of logic, IDS’s provide a restriction on the classical set of models. Therefore, they can be used as an underlying semantics which defines the intended models or, equivalently, the intended conclusions from a given scenario description. In other words, there is a two-way relationship between the IDS ‘model’ and a logic of actions and change. Chapters 1 and 2 addressed this common topic from the IDS point of view; the preSCDt chapter will do it from the point of view of the logic.