An introduction to soft constraints with bipolar preferences
July 23, 12.00 (Aula D)
Palazzo delle Scienze - Cagliari
Dipartimento di Informatica - Università di Pisa
Abstract. Soft constraints refer to a family of formalisms adopted for modelling and optimising problems of a combinatorial nature. They usually take the form of a set of variable assignments, each one of them equipped with a value denoting the degree of preference for that assignment as part of the solution to the problem. Clearly, the more general the algebraic structures chosen for the set of preferences, the better, as long as it is guaranteed that some optimization techniques (transforming a problem into an equivalent, yet simpler to solve) can still be applied. The talk introduces “soft constraints with bipolar preferences”, one of the most general proposal so far, and illustrates its expressiveness as well as the extent to which such techniques can be recovered.