On the logic of iterated belief revision
Artificial Intelligence
Machine Learning
Autonomous Agents and Multi-Agent Systems
Representing Transition Systems by Logic Programs
LPNMR '99 Proceedings of the 5th International Conference on Logic Programming and Nonmonotonic Reasoning
On the revision of preferences and rational inference processes
Artificial Intelligence
Dynamic Epistemic Logic
From Belief Change to Preference Change
Proceedings of the 2008 conference on ECAI 2008: 18th European Conference on Artificial Intelligence
Defining relative likelihood in partially-ordered preferential structures
Journal of Artificial Intelligence Research
An action description language for iterated belief change
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
| Hi-index | 0.00 |
An agent may have preferences over states and an agent may have preferences over actions. In this paper, we explore the connection between these distinct forms of preference, in the context where action effects are given by a transition system. We illustrate that preferences over actions can not always be reduced to preferences over states, even under very general conditions. It is possible, however, to define a natural notion of consistency between the two forms of preference. Moreover, it is possible to precisely specify which preferences over actions can be expressed in terms of preferences over states. We encode preferences over actions in a logic programming framework that allows us to automatically determine when preferences over actions can be reduced to preferences over states. Our framework facilitates the high-level analysis of preferences by making conflicts explicit. We conclude with a general discussion of conflicting preferences, and we suggest some topics for future work.