J. Doyle and M. P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1-3):97--128, May 1991.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....Independently, Y. Shoham, in [37] proposed a general semantics for nonmonotonic reasoning, based on preferences among models. The link between the properties of choice functions studied by Social Choice researchers and Nonmonotonic Reasoning has been put in evidence by Doyle and Wellman [7], Rott [30] and Lindstr om [23] Lindstr om generalizes the nitary framework considered in Revealed Preference Theory to an in nitary framework. In [31, 32] Karl Schlechta rediscovered choice functions and their properties, in the in nitary framework. He also considered an additional property ....

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. Arti cial Intelligence, 49(1-3):97-128, May 1991.

....has no simple answer and has probably no unique answer good for everyone in every situation. It may well be the case that, in different situations or for different domains of knowledge, the pragmatically right answers to the question of the title differ. This feeling has been recently expressed in [9]. The first part of this paper defines the notion of a rational set of assertions and defends the thesis that any reasonable answer to the question of the title must consist of such a set of assertions. Thesis 1 The set of assertions entailed by any set of assertions is rational. The second part ....

Jon Doyle and Michael P. Wellman. Impediments to universal preferencebased default theories. Artificial Intelligence, 49(1--3):97--128, May 1991.

....decision making, which should be of interest to researchers in qualitative decision theory. 22 Although researchers in arti cial intelligence (and to some extent statisticians) have examined and used various qualitative notions of knowledge, belief (e.g. 12, 29, 8] and preference (e.g. [15]) most e ort has been on de ning these concepts and understanding the process of (qualitative) belief revision following new information. More recently, ideas on how qualitative notions of belief and preference can be combined have been examined [44, 5] However, investigations of the foundations ....

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. In Ron Brachman and Hector Levesque, editors, Proc. of First Intl. Conf. on Principles of Knowledge Representation and Reasoning, pages 94-102, Toronto, Canada, May 1989. Morgan Kaufmann.

....designed. Society metaphors have been proposed in the AI literature also in contexts which differ from the artificial social systems setting. Minsky uses a society metaphor in his work on the society of mind [36] The notion of social choice is an important element in e.g. the work of Jon Doyle [15]. Finally, social metaphors appear 40 also in the works of Fox, Kornfeld and Hewitt, Malone, and Simon ( 20] 27] 34] 57] concerning organization theory. In this paper we treat the notion of an artificial social system in a relatively narrow sense, and with a particular point of view in ....

J. Doyle and M.P. Wellman. Impediments to Universal Preference-Based Default Theories. In Proceedings of the 1st conference on principles of knowledge representation and reasoning, 1989.

....decision making, which should be of interest to researchers in qualitative decision theory. 22 Although researchers in artificial intelligence (and to some extent statisticians) have examined and used various qualitative notions of knowledge, belief (e.g. 12, 29, 8] and preference (e.g. [15]) most effort has been on defining these concepts and understanding the process of (qualitative) belief revision following new information. More recently, ideas on how qualitative notions of belief and preference can be combined have been examined [44, 5] However, investigations of the ....

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. In Ron Brachman and Hector Levesque, editors, Proc. of First Intl. Conf. on Principles of Knowledge Representation and Reasoning, pages 94--102, Toronto, Canada, May 1989. Morgan Kaufmann.

....into descriptive and prescriptive approaches. In the former case, one has a wish list where the intent is that one way or another the highest ranked rules be applied. In the latter case the ordering reflects the order in which rules should be applied. The relationship of Doyle and Wellman s work [Doyle and Wellman, 1991] to such preferences is also discussed in [Delgrande and Schaub, 2000a] Rintanen, 1998b] addresses descriptive preference orders in default logic. This notion of preference differs conceptually from the preference orderings dealt with here. In particular, Rintanen s approach is based on a ....

J. Doyle and M. Wellman. Impediments to universal preferencebased default theories. Artificial Intelligence, 49(1-3):97--128, 1991.

....preference orderings induced by the obligations, norms and desires of the agent. As said before, these preference orderings have to be combined into one ordering on possible goals. In order to achieve this it would be nice to have a general, intuitive mechanism. Unfortunately, social choice theory [1, 8] points out that it is not possible to find such an aggregation mechanism if it has to possess a number of intuitive properties. These properties are: collective rationality. The aggregate preference ordering is a function of the separate preference orderings. Pareto principle. If i ....

Doyle, J. and Wellman, M. Impediments to Universal Preference-Based Default Theories, Artifical Intelligence, Vol.49, Nr.1-3, pages 97-128, 1991.

....preference, canceling preferences, preferences among preferences, and preferences by default. As well, elsewhere we show that we can capture the approach of [Brewka and Eiter, 2000] It might be argued that, given complete information about preferences, such generality may not be required: [Doyle and Wellman, 1991], building on work by Arrow, argue that in any preference based default theory, for coherence, one requires a dictator to adjudicate preferences. That is, in a complete system there must be, essentially, some way of determining a unique, complete, priority ordering. So in this sense, all one ....

J. Doyle and M.P. Wellman. Impediments to universal preference -based default theories. Artificial Intelligence, 49(1-3):97--128, 1991.

....Independently, Y. Shoham, in [37] proposed a general semantics for nonmonotonic reasoning, based on preferences among models. The link between the properties of choice functions studied by Social Choice researchers and Nonmonotonic Reasoning has been put in evidence by Doyle and Wellman [7], Rott [30] and Lindstrom [23] Lindstrom generalizes the finitary framework considered in Revealed Preference Theory to an infinitary framework. In [31, 32] Karl Schlechta rediscovered choice functions and their properties, in the infinitary framework. He also considered an additional property ....

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1--3):97--128, May 1991.

.... THE RELATIONSHIP BETWEEN ABDUCTION AND DEDUCTION L uca CONSOLE , Daniele THESEIDER DUPRE, Pietro TORASSO Dipartimento di Informatica Universita di Torino 1 Corso Svizzera 185 0149 Torino (Italy) 2 Phone number : 39) 11 771200 Fax number : 39) 11 751603 E Mail: lconsole di.unito.it dtd di.unito.it t A torasso di.unito.i BSTRACT n a The aim of this paper is at analyzing from various points of view the relationships betwee bduction and deduction. In ....

....single out the preferred explanation to a problem or to rank alternaa tive explanations. Many researchers suggest that such a problem is just a matter of pragmatics nd that no logical (domain and task independent) criterion can be used to support the choice k (see, for example, the comments in [14, 22, 33] In the object level framework, when no nowledge about abducible atoms is available (as we have assumed in this section) it is t natural to introduce the criterion of minimal information as a partial criterion to compare solu ions. In other words, we prefer those explanations which ....

[Article contains additional citation context not shown here]

Doyle, J. and Wellman, M., "Impediments to Universal Preference-Based Default Theories," pp. , T 94-102 in Proc. 1st Conference on Principles of Knowledge Representation and Reasoning oronto (1989).

....and Morgenstern is the basis of utility theory, which by itself is the basis of decision theory and mathematical economics. It is well known that there are tight connections between these areas and fundamental aspects of AI, such as reasoning about uncertainty [25] nonmonotonic reasoning [6], learning [3] and multi agent systems [9] In their work, Von Neumann and Morgenstern assume an agent has a set of possible outcomes to consider, and a set of preferences on these outcomes. The agent s preferences are assumed to obey some natural requirements or postulates. We will not discuss ....

J. Doyle and M.P. Wellman. Impediments to Universal Preference-Based Default Theories. In Proceedings of the 1st conference on principles of knowledge representation and reasoning, 1989.

....obtains. As was said, Doyle derives a dictatorial result, which implies that Omega is an ultrafilter. But as was said Omega forms a filter in the case of individual partial orders. So Doyle s Impossibility Theorem means that there are no proper filters i.e. that each filter is an ultrafilter [7]. This does not mean that partial orders can be in every case be aggregated in a weak order. Instead it means that in most cases a partial global order of extensions will obtain. So the possibility of an universal reasoner remains open, a rational entity provided by the aggregation of rational ....

Doyle, J. and Wellman, M.P. 1991. Impediments to universal preferencebased default default theories. Artificial Intelligence 49(1), May, 97-128.

....First, we believe that identifying the connection between CF and Social Choice theory allows CF researchers to leverage a great deal of previous work on preference and utility aggregation. A Social Choice perspective on combining default reasoning rules has yielded valuable insights for that task [Doyle and Wellman, 1991], and similar benefits may accrue for cf. Additionally, weighted versions of any of the many proposed voting schemes [Fishburn, 1973] are immediate candidates for new CF algorithms. Understanding what is theoretically impossible is an important first step in algorithm design. We believe that the ....

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49: 97--128, 1991.

....systems. The core idea is that a set of heuristics may behave like a set of individuals who have to vote and that finding the best solution in the sense of a group of possibly antagonistic rules is analogous to a social choice problem. Similar considerations can be found in very recent works by Doyle and Wellman (1991) for the choice of an extension in default reasoning. They come up with negative results regarding the possibility of improving over the rigid lexicographic priority mechanisms that have been proposed for conflict resolution. In our work we try to cope with these difficulties using fuzzy set ....

....a way to aggregate results coming from various points of view that satisfy these five properties is vain. It will consist in selecting a particular point of view and rejecting the others if one takes into account only the ranking induced on the decisions. Similar conclusions are emphasized by Doyle and Wellman (1991) in non monotonic reasoning. So as to avoid the conclusion of Arrow s theorem, one may want either to abandon or weaken one or several properties and find a voting procedure compatible with this new set of axioms. Unfortunately, relaxing the initial properties make the voting procedures ....

Doyle J., Welman M.P. (1991) Impediments to universal preference-based default theories. In: Knowledge Representation, Special Issue of Artificial Intelligence.

No context found.

J. Doyle and M. P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1-3):97--128, May 1991.

....On the other hand, finite sets of goals provide a focus for reasoning and action in a way that specifying utilities for the infinity of conceivable circumstances does not. To gain the advantages of both goals and preferences, one must introduce some new notion to connect the two. See (Wellman and Doyle, 1991) for an approach using multi attribute descriptions of outcomes to define goals as conditions preferred to their opposites, holding other things equal. 2.3 Decision theory Most work in artificial intelligence that makes use of economic rationality draws on the specific theory of subjective ....

....no fixed upper limit to the time available. Different tasks will have different deadlines in different situations (for example, emergency room medical diagnosis has much shorter deadlines than diagnosis in the doctor s office) and deadlines can sometimes be postponed. More generally, the 10 In (Doyle and Wellman, 1991), however, these multiple extensions are viewed as stemming from inconsistent preferences about belief. 13 Doyle reasoner may be able to trade other resources at its disposal for more time (for example, processor speeds might be changed if increased error rates are acceptable) Computational ....

[Article contains additional citation context not shown here]

Doyle, J. and Wellman, M. P. 1991. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1-3):97--128.

....as mere failures in performance. In fact, rational representation is just one element of a rich interplay between the various forms of rationality in representation and reasoning. A comprehensive treatment of these forms and their interactions is beyond the scope of this paper, but is pursued in (Doyle 1988a) The result is a theory of bounded rationality in which limited forms of rationality are themselves used to shape the overall limits on the agent s rationality. 1 The sense in which we use represent here is distinct from the sense in which the agent s beliefs represent something about the ....

....rules and the conflicting preferences they embody. We cannot pursue the point here, but reasoning with conflicting default rules turns out to be formally the same problem as decision making or self government by groups of people, thus justifying the second sense of the term representation. See (Doyle and Wellman 1988) for more on this connection between rational knowledge representation and theories of social or political decision making. 6 Rational expenditure of limited resources If an agent s manifest beliefs and derivation rules are logically consistent and complete, then the consequences of these ....

[Article contains additional citation context not shown here]

Doyle, J., and Wellman, M. P., 1988. Impediments to universal preference-based default theories, Proceedings of the First International Conference on Principles of Knowledge Representation and Reasoning, San Mateo, CA: Morgan Kaufmann.

No context found.

Doyle, J., and Wellman, M. P. 1991. Impediments to Universal Preference-Based Default Theories. Artificial Intelligence 49(1--3):97--128.

No context found.

J. Doyle and M.P. Wellman. Impediments to Universal Preference-Based Default Theories. In Proceedings of the 1st conference on principles of knowledge representation and reasoning, 1989.

No context found.

J. Doyle and M. P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49, 97--128, 1991.

No context found.

J. Doyle and M.P. Wellman. Impediments to universal preference -based default theories. Artificial Intelligence, 49(1-3):97--128, 1991.

No context found.

J. Doyle and M. P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1--3), 1992.

No context found.

J. Doyle and M. Wellman. Impediments to Universal Preference-Based Default Theories, Artifical Intelligence, Vol.49, Nr.1-3, pages 97-128, 1991.

No context found.

J. Doyle and M. P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49(1--3):97--128, 1992.

No context found.

Jon Doyle and Michael P. Wellman. Impediments to universal preference-based default theories. Artificial Intelligence, 49: 97--128, 1991.

First 50 documents

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC