Applications of transaction logic to knowledge representation

Abstract

We present applications of the recently proposed Transaction Logic — an extension of classical logic that accounts in a clean and declarative fashion for the phenomenon of state changes in knowledge bases, including logic programs and databases. Transaction Logic has a natural model theory and a sound- and-complete proof theory, but unlike many other logics, it allows users to program transactions. Its semantics also leads naturally to features whose amalgamation in a single logic has proved elusive in the past. These features include both hypothetical and committed updates, static and dynamic constraints, nondeterministic actions, nested transactions, bulk updates, view updates, active databases, subjunctive queries, and more. Finally, Transaction Logic holds promise as a logical model of hitherto non-logical phenomena, including procedural knowledge in AI, and the behavioral aspect of object-oriented databases, especially methods with side effects. This paper outlines the model theory of Transaction Logic, and then focuses on some of its applications to AI, including action definition and execution, planning, and dynamic constraints.