Proposta de dissertação do MEI Título: Interval Quadrature over Continuous Constraints Proponente(s): Jorge Cruz Créditos: 42 ECTS Área científica: Decision Support and Artificial Intelligence Início preferencial: Qualquer semestre URL: Já estão em curso trabalhos preliminares executados pelo alunos: Breve descrição: This MSc proposal addresses the issue of improving the efficacy of interval quadrature methods in the context of the branch-and-prune algorithm of continuous constraint programming. The combination of safe and probabilistic reasoning is of paramount importance for the support of sustained decisions on several science and engineering problems. The probabilistic constraint framework computes the probability of a constraint event, based on the efficiency of constraint programming to get a close box covering of the region of integration, and on the efficacy of the quadrature methods to calculate the integrals over the obtained boxes. The innovative idea to be developed in the MSc work is to take advantage of the covering box partition to compute sharp reliable enclosures of the event probability value by applying interval methods for verified quadrature that explore the properties of the constraints on certain regions to rewrite the constraint set and simplify the computation. Observações: