Logics for AI

Logics for AI Contact: John-Jules Meyer Logic is a recognised tool for AI research in reasoning, knowledge representation and communication. It provides semantics for notations, and methods for automating inference.
PEOPLE The members of the Logics for AI group PUBLICATIONS Publications on logics for AI	Research topics within the group include: Modal logic Epistemic logic Nonmonotonic logic Belief revision Deontic logic
Modal logic Modal Logic is one of the 'classical' logics for AI. It is useful for modelling reasoning about knowledge, actions, time or obligations.
Epistemic logic Epistemic logics apply the techniques of modal logic to reasoning about knowledge. Both individual and group knowledge is studied. The study of epistemic logic is relevant for, for instance, (specification of) communication protocols and cooperation.
Nonmonotonic logic Nonmonotonic logics formalise unsound but reasonable patterns of reasoning with uncertain, incomplete and inconsistent information. Our group has done work on default logic, circumscription, modal nonmonotonic logics, and more recently on logics for defeasible argumentation. Key publications: Prakken & Sartor, 1997, Vreeswijk, 1997.
Belief revision Updating one's beliefs is not trivial. Belief revision studies how to revise one's knowledge if a fact is discovered that is inconsistent with it.
Deontic logic Deontic logic formalises normative modalities, such as 'obligatory', 'permitted' and 'forbidden'. Deontic logic can be applied to representation of normative (e.g. legal) knowledge, and to the modelling of norm-governed agent or system interaction. Key publications: Prakken & Sergot, 1996.

[IS home page] [People] [Publications] [ICS home page] [UU home page]