If I don’t know, should I infer? Reasoning around ignorance in a many-valued framework
Résumé
—Many-valued logic allows to reason with partial truth measured by degrees on a discrete scale, but it suffers from an ambiguous interpretation of the middle truth level, considered as intermediate truth or as ignorance, i.e. inability to assess truth. The L e M extension introduces an additional value, outside the truth scale, to distinguish between these two notions. This paper studies L e M from a logical perspective, examining how to reason in this framework: it discusses the definition of appropriate semantics for the logical connectives and it considers an inference task, proposing a Modus Ponens variant for L e M .
Domaines
Intelligence artificielle [cs.AI]Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...