The Experts below are selected from a list of 3 Experts worldwide ranked by ideXlab platform
Dupré D - One of the best experts on this subject based on the ideXlab platform.
-
Representing ramifications in an event-based language
Department of Computer Science K.U.Leuven Leuven Belgium, 1997Co-Authors: Van Belleghem Kristof, Denecker Marc, Dupré DAbstract:In the last couple of years, several high-level languages have been proposed for modeling Actions and change, following the example of Gelfond and Lifschitz's ${\cal A}$ language. The goal of these languages is to analyse in a simplified context the fundamental issues in reasoning about Actions and the constructs required for dealing with them. In this paper we present a narrative-based language ${\cal ER}$ with a linear time structure, which is designed to deal with general fluent and change dependencies, in particular ramifications, both of consecutive and simultaneous Actions. We argue that a combination of state constraints, effect propagation rules (causal laws) and Action Precondition rules is necessary to correctly represent all such dependencies. In particular, we introduce causal laws stating that a change is triggered by a change in truth value of a {\em complex fluent formula}; we show such laws to be useful for a compact representation of ramifications, including in a natural way ramifications of simultaneous Actions.We motivate and define a constructive semantics for simple and complex causal laws based on the principle of inductive definitions. We design ${\cal ER}$ in such a way that it is able to deal flexibly with complete and incomplete knowledge on Action occurrences, Action ordering and the initial state of the world. We present a mapping of ${\cal ER}$ to Open Logic Programming using the Event Calculus, and prove the correctness of this mapping. As the mapping is rather straightforward, ${\cal ER}$ can in a sense be considered a language for analysing the underlying principles of the Event Calculus in Open Logic Programming. Further on, we extend the basic ${\cal ER}$ language with constructs for dealing with nondeterminism and delayed ramifications, and extend the mapping to Open Logic Programming accordingly. Finally we indicate how influence information, as introduced by Thielscher, can be used for derivingcausal laws from state constraints (semi-)automatically in the ${\cal ER}$ setting. We discuss how our approach compares with and differs from Thielscher's. Moreover we compare the ${\cal ER}$ approach to the ramification problem with many recent approaches.status: publishe
Van Belleghem Kristof - One of the best experts on this subject based on the ideXlab platform.
-
Representing ramifications in an event-based language
Department of Computer Science K.U.Leuven Leuven Belgium, 1997Co-Authors: Van Belleghem Kristof, Denecker Marc, Dupré DAbstract:In the last couple of years, several high-level languages have been proposed for modeling Actions and change, following the example of Gelfond and Lifschitz's ${\cal A}$ language. The goal of these languages is to analyse in a simplified context the fundamental issues in reasoning about Actions and the constructs required for dealing with them. In this paper we present a narrative-based language ${\cal ER}$ with a linear time structure, which is designed to deal with general fluent and change dependencies, in particular ramifications, both of consecutive and simultaneous Actions. We argue that a combination of state constraints, effect propagation rules (causal laws) and Action Precondition rules is necessary to correctly represent all such dependencies. In particular, we introduce causal laws stating that a change is triggered by a change in truth value of a {\em complex fluent formula}; we show such laws to be useful for a compact representation of ramifications, including in a natural way ramifications of simultaneous Actions.We motivate and define a constructive semantics for simple and complex causal laws based on the principle of inductive definitions. We design ${\cal ER}$ in such a way that it is able to deal flexibly with complete and incomplete knowledge on Action occurrences, Action ordering and the initial state of the world. We present a mapping of ${\cal ER}$ to Open Logic Programming using the Event Calculus, and prove the correctness of this mapping. As the mapping is rather straightforward, ${\cal ER}$ can in a sense be considered a language for analysing the underlying principles of the Event Calculus in Open Logic Programming. Further on, we extend the basic ${\cal ER}$ language with constructs for dealing with nondeterminism and delayed ramifications, and extend the mapping to Open Logic Programming accordingly. Finally we indicate how influence information, as introduced by Thielscher, can be used for derivingcausal laws from state constraints (semi-)automatically in the ${\cal ER}$ setting. We discuss how our approach compares with and differs from Thielscher's. Moreover we compare the ${\cal ER}$ approach to the ramification problem with many recent approaches.status: publishe
Denecker Marc - One of the best experts on this subject based on the ideXlab platform.
-
Representing ramifications in an event-based language
Department of Computer Science K.U.Leuven Leuven Belgium, 1997Co-Authors: Van Belleghem Kristof, Denecker Marc, Dupré DAbstract:In the last couple of years, several high-level languages have been proposed for modeling Actions and change, following the example of Gelfond and Lifschitz's ${\cal A}$ language. The goal of these languages is to analyse in a simplified context the fundamental issues in reasoning about Actions and the constructs required for dealing with them. In this paper we present a narrative-based language ${\cal ER}$ with a linear time structure, which is designed to deal with general fluent and change dependencies, in particular ramifications, both of consecutive and simultaneous Actions. We argue that a combination of state constraints, effect propagation rules (causal laws) and Action Precondition rules is necessary to correctly represent all such dependencies. In particular, we introduce causal laws stating that a change is triggered by a change in truth value of a {\em complex fluent formula}; we show such laws to be useful for a compact representation of ramifications, including in a natural way ramifications of simultaneous Actions.We motivate and define a constructive semantics for simple and complex causal laws based on the principle of inductive definitions. We design ${\cal ER}$ in such a way that it is able to deal flexibly with complete and incomplete knowledge on Action occurrences, Action ordering and the initial state of the world. We present a mapping of ${\cal ER}$ to Open Logic Programming using the Event Calculus, and prove the correctness of this mapping. As the mapping is rather straightforward, ${\cal ER}$ can in a sense be considered a language for analysing the underlying principles of the Event Calculus in Open Logic Programming. Further on, we extend the basic ${\cal ER}$ language with constructs for dealing with nondeterminism and delayed ramifications, and extend the mapping to Open Logic Programming accordingly. Finally we indicate how influence information, as introduced by Thielscher, can be used for derivingcausal laws from state constraints (semi-)automatically in the ${\cal ER}$ setting. We discuss how our approach compares with and differs from Thielscher's. Moreover we compare the ${\cal ER}$ approach to the ramification problem with many recent approaches.status: publishe
