The Experts below are selected from a list of 21012 Experts worldwide ranked by ideXlab platform
Michael Thielscher - One of the best experts on this subject based on the ideXlab platform.
-
The Qualification Problem: A solution to the Problem of anomalous models
2015Co-Authors: Michael ThielscherAbstract:Abstract. Intelligent agents in open environments inevitably face the Quali¯cation Problem: The executability of an action can never be predicted with absolute certainty; unexpected cir-cumstances, albeit unlikely, may at any time prevent the successful performance of an action. Reasoning agents in real-world environments rely on a solution to the Quali¯cation Problem in order to make useful predictions but also to explain and recover from unexpected action failures. Yet the main theoretical result known today in this context is a negative one: While a solution to the Quali¯cation Problem requires to assume away by default abnormal quali¯cations of actions, straightforward minimization of abnormality falls prey to the production of anomalous models. We present an approach to the Quali¯cation Problem which resolves this anomaly. Anomalous models are shown to arise from ignoring causality, and they are avoided by appealing to just this concept. Our theory builds on the established predicate logic formalism of the Fluent Cal-culus as a solution to the Frame Problem and to the Rami¯cation Problem in reasoning about actions. The monotonic Fluent Calculus is enhanced by a default theory in order to obtain the nonmonotonic approach called for by the Quali¯cation Problem. The approach has been implemented in an action programming language based on the Fluent Calculus and successfully applied to the high-level control of robots
-
a language for default reasoning about actions
Correct Reasoning, 2012Co-Authors: Hannes Strass, Michael ThielscherAbstract:Action languages allow for a concise representation of actions and their effects while at the same time being easily readable and writable for humans. In this paper, we introduce ${\mathcal D}$, the first action language that centres around default reasoning about actions and change. It allows to specify normal Reiter-style defaults and provides a semantics that has at its core the groundedness of conclusions. This is not only of use for default conclusions, but also for ${\mathcal D}$'s solution to the ramification Problem. Additionally, our language does not suffer from Yale Shooting-like counterexamples since it uses different mechanisms for default persistence and default change. The answer set programming paradigm is used as the basis of ${\mathcal D}$'s implementation, which we prove sound and complete with respect to the language's semantics. We finally present a showcase application for the language: its straightforward solution to the Qualification Problem.
-
Addressing the Qualification Problem in FLUX
2001Co-Authors: Yves Martin, Michael ThielscherAbstract:The Qualification Problem arises for planning agents in realworld environments, where unexpected circumstances may at any time prevent the successful performance of an action. We present a logic programming method to cope with the Qualification Problem in the action programming language Flux, which builds on the Fluent Calculus as a solution to the fundamental Frame Problem. Our system allows to plan under the default assumption that actions succeed as they normally do, and to reason about these assumptions in order to recover from unexpected action failures
-
Causality and the Qualification Problem
Morgan Kaufmann, 1996Co-Authors: Michael ThielscherAbstract:In formal theories for reasoning about actions, the Qualification Problem denotes the Problem to account for the many conditions which, albeit being unlikely to occur, may prevent the successful execution of an action. By a simple counter-example in the spirit of the well-known Yale Shooting scenario, we show that the common straightforward approach of globally minimizing such abnormal disQualifications is inadequate as it lacks an appropriate notion of causality. To overcome this difficulty, we propose to incorporate causality by treating the proposition that an action is qualified as a fluent which is initially assumed away by default but otherwise potentially indirectly affected by the execution of actions. Our formal account of the Qualification Problem includes the proliferation of explanations for surprising disQualifications and also accommodates so-called miraculous disQualifications. We moreover sketch a version of the fluent calculus which involves default rules to address ..
-
Qualification and Causality
1996Co-Authors: Michael ThielscherAbstract:In formal theories for reasoning about actions, the Qualification Problem denotes the Problem to account for the many conditions which, albeit being unlikely to occur, may prevent the successful execution of an action. While a solution to this Problem must involve the ability to assume away by default these abnormal disQualifications of actions, the common straightforward approach of globally minimizing them is inadequate as it lacks an appropriate notion of causality. This is shown by a simple counter-example closely related to the well-known Yale Shooting scenario. To overcome this difficulty, we propose to incorporate causality by treating the fact that an action is qualified as ordinary fluent, i.e., a proposition which may change its truth value in the course of time by potentially being (indirectly) affected by the execution of actions. Abnormal disQualifications then are initially assumed away, unless there is evidence to the contrary. Our formal account of the Qualification pro..
Schubert, Lenhart K. - One of the best experts on this subject based on the ideXlab platform.
-
Explanation Closure, Action Closure, and the Sandewall Test Suite for Reasoning about Change
University of Rochester. Computer Science Department., 2012Co-Authors: Schubert, Lenhart K.Abstract:Explanation closure (EC) axioms were previously introduced as a means of solving the frame Problem. This paper provides a thorough demonstration of the power of EC combined with action closure (AC) for reasoning about dynamic worlds, by way of Sandewall's test suite of 12-or-so Problems [Sandewall 1991; 1992]. Sandewall's Problems range from the "Yale turkey shoot" (and variants) to the "stuffy room" Problem, and were intended as a test and challenge for nonmonotonic logics of action. The EC/AC-based solutions for the most part do not resort to nonmonotonic reasoning at all, yet yield the intuitively warranted inferences in a direct, transparent fashion. While there are good reasons for ultimately employing nonmonotonic or probabilistic logics---e.g., pervasive uncertainty and the Qualification Problem---this does show that the scope of monotonic methods has been underestimated. Subsidiary purposes of the paper are to clarify the intuitive status of EC axioms in relation to action effect axioms; and to show how EC, previously formulated within the situation calculus, can be applied within the framework of a temporal logic similar to Sandewall's "discrete fluent logic," with some gains in clarity
Lenhart K Schubert - One of the best experts on this subject based on the ideXlab platform.
-
explanation closure action closure and the sandewall test suite for reasoning about change
Journal of Logic and Computation, 1994Co-Authors: Lenhart K SchubertAbstract:Explanation closure (EC) axioms were previously introduced as a means of solving the frame Problem. This paper provides a thorough demonstration of the power of EC combined with action closure (AC) for reasoning about dynamic worlds, by way of Sandewall’s test suite of 12-or-so Problems [29], [30], [31]. Sandewall’s Problems range from the “Yale turkey shoot” (and variants) to the “stuffy room” Problem, and were intended as a test and challenge for nonmonotonic logics of action. The EC/AC-based solutions for the most part do not resort to nonmonotonic reasoning at all, yet yield the intuitively warranted inferences in a direct, transparent fashion. While there are good reasons for ultimately employing nonmonotonic or probabilistic logics - e.g., pervasive uncertainty and the Qualification Problem - this does show that the scope of monotonic methods has been underestimated. Subsidiary purposes of the paper are to clarify the intuitive status of EC axioms in relation to action effect axioms; and to show how EC, previously formulated within the situation calculus, can be applied within the framework of a temporal logic similar to Sandewall’s “discrete fluent logic”, with some gains in clarity.
Yves Martin - One of the best experts on this subject based on the ideXlab platform.
-
Specifying Explanations for an Action Qualification with Sensing Actions
2009Co-Authors: Yves MartinAbstract:Abstract. Planning agents in real-world environments have to face the Qualification Problem, i.e., the failure of an action execution due to unexpected circumstances. Sensing actions are used to derive additional state knowledge. We present a high-level programming method to combine these two approaches in the agent programming language FLUX, which builds on the general action representation formalism of the Fluent Calculus. It is shown how this combination allows for an efficient reasoning about the causes of unexpected action failures. The method is especially indicated if there are many possible reasons for a failure which have the same prior probabilities. The determined explanation for an action failure then helps an agent to recover from it.
-
Addressing the Qualification Problem in FLUX
2001Co-Authors: Yves Martin, Michael ThielscherAbstract:The Qualification Problem arises for planning agents in realworld environments, where unexpected circumstances may at any time prevent the successful performance of an action. We present a logic programming method to cope with the Qualification Problem in the action programming language Flux, which builds on the Fluent Calculus as a solution to the fundamental Frame Problem. Our system allows to plan under the default assumption that actions succeed as they normally do, and to reason about these assumptions in order to recover from unexpected action failures
Erik T Mueller - One of the best experts on this subject based on the ideXlab platform.
-
chapter 12 default reasoning
Commonsense Reasoning, 2006Co-Authors: Erik T MuellerAbstract:Commonsense reasoning requires default reasoning. As one engage in commonsense reasoning, one makes certain assumptions so that he can proceed. This chapter discusses default reasoning in general, temporal default reasoning, and then temporal default reasoning. Further, it also describes a general method for default reasoning about time based on the circumscription of abnormality predicates and how this method can be used to address the Qualification Problem. Then, explains the representation of default events and properties. The event calculus assumes by default that unexpected events do not occur and events have no unexpected effects. In addition, default reasoning can be used to address the Qualification Problem. Default reasoning provides a more elaboration-tolerant way of specifying Qualifications that prevent events from having their intended effects.
