BREWKA G., "Reasoning about Priorities in Default Logic", Proceedings of the AAAI National Conference on Artificial Intelligence, vol. 2, The AAAI Press/The MIT Press, 1994, p. 940-945.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....negation, is potentially quite conceptually confusing, again especially for relatively non technical end users. For inspiration in grappling with these challenges, we have drawn on the idea of prioritized defaults (e.g. 30 ] 27 ] 14 ] 17 ] 15 ] ch. 2 includes a literature review) and [ 3 ] (also includes a literature review) from the field of general non monotonic reasoning. As has been explored there (and in the related knowledge representation and common sense reasoning literatures) it is often easier and more natural, especially for non technical end users, to specify rules in ....

....set of conclusions) This helps provide a conceptually simple semantics, which facilitates understandability, especially by non technical end users. By contrast, many expressively powerful non monotonic reasoning formalisms have multiple extensions (e.g. Prioritized Default Logic variants; see [ 3 ] for review) Fifth, reassuringly, the (acyclic) courteous logic programs class includes (acyclic) consistent extended logic programs as a sub class, both syntactically and semantically. Acyclic) general logic programs are essentially a sub class of consistent extended logic programs (the ....

[Article contains additional citation context not shown here]

Gerhard Brewka. Reasoning about priorities in default logic. In Proceedings of the Twelfth National Conference on Artificial Intelligence (AAAI-94), pages 940--945, Menlo Park, CA / Cambridge, MA, 1994. AAAI Press / MIT Press.

.... for logic programming with negation as failure [Przymusinski, 1988] and terminological logics [Quantz and Royer, 1992] possibilistic logic [Dubois and Prade, 1988] syntax based belief revision formalisms, e.g. Nebel, 1989] and a variety of others, e.g. Brewka, 1989a] Brewka, 1989b] [Brewka, 1994] [Ginsberg, 1988] Zadrozny, 1987] Pollock, 1987] Konolige, 1988] Ryan, 1992b] Ryan, 1992a] Hunter, 1994] From a practical viewpoint, two very important kinds of non monotonic reasoning, in use even before the knowledge representation (KR) field of non monotonic logic started in the late ....

Gerhard Brewka. Reasoning about priorities in default logic. In Proceedings of the Twelfth National Conference on Artificial Intelligence (AAAI-94), pages 940--945, Menlo Park, CA / Cambridge, MA, 1994. AAAI Press / MIT Press.

....has been taken by Konolige [25] in using autoepistemic logic to reason with speci city. He de ned hierarchical autoepistemic theories in which a preference order between sub theories and a syntactical condition on the sub theories ensure that higher priority conclusions will be concluded. Brewka [6] in de ning prioritized default logic also adds a preference order between defaults into a Reiter s default theory and modi es the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [1] develops prioritized default ....

....defaults into a Reiter s default theory and modi es the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [1] develops prioritized default logic to handle speci city in terminological systems. All of the approaches in [6, 1, 32, 27, 25] assume that priorities between defaults are given by the users. Computing speci city is another important issue in approaches to reasoning with speci city. Work from Poole [41] is an early attempt to extract the preference between defaults from the theory. Poole de nes a notion of more speci ....

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the 12th National Conference on Arti cial Intelligence, pages 940-945. AAAI Press, 1994.

....the original language by a set of constants N such that there is a bijective mapping n : D N . We write n # instead of n(#) and we often abbreviate n # i by n i to ease notation) Also, for default # along [5] first suggested naming defaults using a set of aspect functions. See also [8, 1]. with its name n, we sometimes write n : # to render naming explicit. To encode the fact that we deal with a finite set of distinct default rules, we adopt a unique names assertion (UNAN ) and domain closure assertion (DCAN ) with respect to N . So, for a name set N = nk , we add ....

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the AAAI National Conference on Artificial Intelligence, volume 2, pages 940--945. The AAAI Press/The MIT Press, 1994.

....to be a problem, but are instead exploited as a useful feature that can provide an adequate formalization of the multi interpretability often present in real life information. In [14] this feature is expressed by adding as an extra parameter a selection function to a default theory. In [6] and [7] a similar approach is developed, based on priority orderings between defaults. In [8] the notion of belief set operator is introduced to formalize the multiple 21 outcomes of a non monotonic reasoning process, and a selection operator to make a choice between the different options. For the ....

....operator to make a choice between the different options. For the application domain discussed here the latter approach is more suitable, because in this approach first all alternative interpretations are generated, and the selection is made afterwards. In the approaches of [14] 6] and [7] the reasoning process itself is controlled by the selection knowledge in such a manner that only one outcome is generated, and other options remain invisible. Such strategic knowledge is not yet available. However, in the future of this project such strategic knowledge may be acquired so that not ....

[Article contains additional citation context not shown here]

G. Brewka, "Reasoning about Priorities in Default Logic", in: Proceedings of the AAAI-94, 1994.

....to be a problem, but are instead exploited as a useful feature that can provide an adequate formalisation of the multi interpretability often present in real life information. In [14] this feature is expressed by adding as an extra parameter a selection function to a default theory. In [6] and [7] a similar approach is developed, based on priority orderings between defaults. In [8] the notion of belief set operator is introduced to formalise the multiple outcomes of a non monotonic reasoning process, and a selection operator to make a choice between the different options. For the ....

....operator to make a choice between the different options. For the application domain discussed in this paper the latter approach is more suitable, because in this approach first all alternative interpretations are generated, and the selection is made afterwards. In the approaches of [14] 6] and [7] the reasoning process itself is controlled by the selection knowledge in such a manner that only one outcome is generated, and other options remain invisible. Such strategic knowledge is not yet available. However, in the future of this project such strategic knowledge may be acquired so that not ....

[Article contains additional citation context not shown here]

G. Brewka, "Reasoning about Priorities in Default Logic", in: Proceedings of the AAAI-94, 1994.

.... More recently, Brewka published several papers that further elaborate on this idea by introducing default logics that only use normal default rules, and in which the specificity conflicts between default are handled by explicit priority orderings that are added to the logic (see [Br94a] and [Br94b]) We want to generalize this perspective. Selection functions as introduced in [TT92] can be viewed as a generalization of Brewka s explicit priority orderings. The basic intuition is that instead of using justifications as techniques to control the specificity conflicts between defaults ....

G. Brewka, Reasoning about Priorities in Default Logic, Proceedings of the AAAI-94, 1994.

....to be a useful tool for nature conservationists. After multiple interpretations of observation information have been identified, often a choice is made for one of them. Which view is (or which views are) most appropriate presumably requires additional heuristic (strategic) knowledge (cf. 3] [4], 12] One of the 16 areas of future research is to further analyze this choice process, in general terms, but also in particular for the knowledge based system. Future research will focus on the acquisition of this knowledge to be able to support users in the selection process. ....

G. Brewka, "Reasoning about Priorities in Default Logic", in: Proceedings of the AAAI94, 1994.

....is no possibility to perform operations on defaults, producing new defaults. Another di erence is that causal rules are ordered (according to a priority (or prefernce order relation) thus in extensions the minimal causal rule is always preferred. There are known variants of default logic, e.g. [1, 4, 3, 5], where a notion of priority of defaults based on an ordering relation exists. In our wok, we adopt Brewka s formalism for priority in defaut logic [3] In order to describe which sets of formulas can be inferred from a causal theory, we introduce the notion of extension, which is very close to ....

....order relation) thus in extensions the minimal causal rule is always preferred. There are known variants of default logic, e.g. 1, 4, 3, 5] where a notion of priority of defaults based on an ordering relation exists. In our wok, we adopt Brewka s formalism for priority in defaut logic [3]. In order to describe which sets of formulas can be inferred from a causal theory, we introduce the notion of extension, which is very close to default extension. De nition 3 Let B be a set of propositional formulas and c = a causal rule. We say that c can be applied to B i 1: 2 ....

Gerd Brewka. Reasoning about priorities in default logic. In Proceedings of the 11th National Conference on Arti cial Intelligence, pages 940-945. AAAI Press / The MIT Press, 1993.

....has been taken by Konolige [21] in using autoepistemic logic to reason with speci city. He de ned hierarchical autoepistemic theories in which a preference order between sub theories and a syntactical condition on the sub theories ensure that higher priority conclusions will be concluded. Brewka [5] in de ning prioritized default logic also adds a preference order between defaults into a Reiter s default theory and modi es the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [1] develops prioritized default logic ....

....defaults into a Reiter s default theory and modi es the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [1] develops prioritized default logic to handle speci city in terminological systems. All of the approaches in [5, 1, 27, 23, 21] assume that priorities between defaults are given by the users. For this reason, these approaches are sometime called reasoning with explicit speci city. Contrary to approaches to reasoning with explicit speci city are approaches to reasoning with implicit speci city in which a mechanism for ....

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the 12th National Conference on Articial Intelligence, pages 940-945. AAAI Press, 1994.

....has been taken by Konolige [20] in using autoepistemic logic to reason with specificity. He defined hierarchical autoepistemic theories in which a preference order between sub theories and a syntactical condition on the sub theories ensure that higher priority conclusions will be concluded. Brewka [4] in defining prioritized default logic also adds a preference order between defaults into a Reiter s default theory and modifies the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [5] develops prioritized default ....

....defaults into a Reiter s default theory and modifies the semantics of default logic in such a way that guarantees that default of higher priority is preferred. Baader and Hollunder [5] develops prioritized default logic to handle specificity in terminological systems. All of the approaches in [4, 5, 24, 27, 20] assume that priorities between defaults are given by the users. Computing specificity is another important issue in approaches to reasoning with specificity. Work from Poole [31] is an early attempt to extract the preference between defaults from the theory. Poole defines a notion of more ....

Brewka, G.: Reasoning about Priorities in Default Logic, Proc. AAAI'94, Seatle, 1994.

....situation. In [1] preference in extensive disjunctive logic programming is considered. As far as overriding is concerned the technique corresponds rather well with our skeptical defeating, but alternatives are fixed as an atom and its (real) negation. Outside the context of logic programming, [2] proposes to add priorities to the object language of default logic. Extensions are then required to be compatible with this information. OCLP and [2] support different intuitions on the notion of priority, as shown by the following example 10 : Example 17. P 1 : a P 2 : c P 3 : c a P 4 : c ....

....rather well with our skeptical defeating, but alternatives are fixed as an atom and its (real) negation. Outside the context of logic programming, 2] proposes to add priorities to the object language of default logic. Extensions are then required to be compatible with this information. OCLP and [2] support different intuitions on the notion of priority, as shown by the following example 10 : Example 17. P 1 : a P 2 : c P 3 : c a P 4 : c Phi :c 10 For the ease of notation we simply denote the program in our formalism. with P 4 OE P 3 OE P 2 OE P 1 . With our approach, we obtain fa; ....

[Article contains additional citation context not shown here]

Gerhardt Brewka. Reasoning about priorities in default logic. In Barbara Hayes-Roth and Richard Korf, editors, Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 940--945, Menlo Park, California, 1994. American Association for Artificial Intelligence, AAAI Press.

....defaults have to take precedence over desire based defaults in cases when there is any con ict between the two. In the context of the formalism I am proposing, prioritization of belief over desire can be achieved by using a more or less standard account of prioritized defaults. See, for instance, [4]. Technical details are provided in Section 2.4, below. With belief defaults prioritized over desire defaults, we obtain a single extension in Example 2.6, the one generated by the following choices: fRain; Wetg. The next example elaborates the hiking scenario of Example 2.6. Example 2.7. The ....

Gerhardt Brewka. Reasoning about priorities in default logic. In Barbara Hayes-Roth and Richard Korf, editors, Proceedings of the Twelfth National Conference on Articial Intelligence, pages 940{ 945, Menlo Park, California, 1994. American Association for Articial Intelligence, AAAI Press.

....Question answering is based on the semantic representation of the KB, see Section 4) We show in this Section that the question answering may be viewed as a result of prioritized default reasoning. First, some preliminaries (we follow the presentation of [2] 6 The presentation is based on [4]. De nition 30 ( 2] Let a default theory T = S; D) be given. Let = hd 0 ; d 1 ; i be a sequence of defaults d i 2 D. The initial segment of length k is denoted by [k] In( Th(S [ fcons(d) d occurs in g) Out( f: 2 just(d) for some d occuring in g A process is ....

....good world in K sP , then Th(w) is a prioritized extension of T . Theorem 38 Let E be a prioritized extension of the default theory T associated with K sP . Then there is in K sP a good world w s.t. E = Th(w) An introduction of a dynamics into prioritized default reasoning has been studied in [4], too. 7 Conclusions A Kripkean semantics of multidimensional dynamic logic programs has been introduced in the paper. The semantics presents a view on a modular and dynamic knowledge representation, based on MDyLoP. The semantic representation of a KB is changed when the current module is ....

Brewka, G. Reasoning about priorities in default logic. Proc. of the AAAI94

....OE n theatre . Finally, the news tells us that the theatre is closed, closed (theatre) this blocks default rule ffi theatre and we only conclude goto(night club ) as above, despite the presence of n night club OE n theatre . Next, we consider an example from [Gordon, 1993] discussed in [Brewka, 1994b] A person wants to find out if her security interest in a certain ship is perfected , or legally valid. This person has possession of the ship, but has not filed a financing statement. According to the code UCC, a security interest can be perfected by taking possession of the ship. However, ....

....ucc) All other instances of these axioms are eliminated by deriving x Sigma y. We then conclude by ls (ucc; sma) that ucc OE sma. This blocks lp(sma; ucc) since its justification sma OE ucc has become refuted. Thus, sma Sigma ucc 2 E yielding ok(sma) and subsequently :perfected . Brewka, 1994b] solves this problem by first generating 4 entire extensions, where E 1 fperfected ; sma OE uccg, E 2 f:perfected ; sma OE uccg, E 3 fperfected ; ucc OE smag, E 4 f:perfected ; ucc OE smag. In a second step he rules out E 1 ; E 2 ; E 3 since they do not verify a certain priority ....

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the AAAI National Conference on Artificial Intelligence, volume 2, pages 940--945. The AAAI Press/The MIT Press, 1994.

No context found.

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the AAAI National Conference on Artificial Intelligence, volume 2, pages 940--945. The AAAI Press/The MIT Press, 1994.

....The full version of this paper was published in Journal of Arti cial Intelligence Research, 4 (1996) 19 36. 1 What we would like to have, therefore, is an approach that allows us to represent preference information in the language and derive such information dynamically 1 . In a recent paper [4] the author has described a variant of normal default logic in which reasoning about preferences is possible. Although the version of default logic presented in this earlier paper produces reasonable results in most cases, this approach has several drawbacks: 1. The approach is computationally ....

....b of r b 62 Cl(RX n DomX;R i 1 (r) or b 0 2 X, where b 0 = b if b is an atom and a if b = a. 9 Now n 2 2 SAFE pr S1 (P 2 ) since n 2 dominates n 1 wrt. S 1 and the empty set of rules. We thus conclude :b as intended. The least xpoint is S 2 = fn 2 n 1 ; n 1 n 2 ) bg In [4] we used an example to illustrate the possible non existence of extensions in our earlier approach. This example involved two normal defaults each of which had the conclusion that the other one is to be preferred. The prioritized logic programming representation of this example is the following: ....

[Article contains additional citation context not shown here]

Brewka, Gerhard, Reasoning About Priorities in Default Logic, Proc. AAAI-94, Seattle, 1994

....with respect to general principles which, as we argue, any prioritized variant of default logic should satisfy. For space reasons, a detailed comparison of our approach to the many other variants of prioritized default logic is necessarily superficial. Rintanen s approach and the approaches in [16, 5, 1, 6] have already been briefly mentioned. The latter handle priorities such that in the (re)construction of an extension, only some of the applicable defaults can be fired in each step. In [9] priorities are handled by encoding them into the object level rather than constraining the construction of ....

G. Brewka. Reasoning About Priorities in Default Logic. Proc. AAAI '94, 1994.

No context found.

BREWKA G., "Reasoning about Priorities in Default Logic", Proceedings of the AAAI National Conference on Artificial Intelligence, vol. 2, The AAAI Press/The MIT Press, 1994, p. 940-945.

No context found.

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the AAAI National Conference on Artificial Intelligence, volume 2, pages 940--945. AAAI Press/The MIT Press, 1994.

No context found.

G. Brewka. Reasoning about priorities in default logic. In AAAI-94, 1994.

No context found.

G. Brewka. Reasoning about priorities in default logic. In Proceedings of the AAAI National Conference on Artificial Intelligence, volume 2, pages 940--945. The AAAI Press/The MIT Press, 1994.

No context found.

G. Brewka. Reasoning about priorities in default logic. In Proceedings AAAI'94, pages 940--945. The AAAI Press, 1994.

No context found.

G Brewka. Reasoning about priorities in default logic. In Proceedings of the Twelfth National Conference on Artificial Intelligence (AAAI'94), pages 420--425. MIT Press, 1994.

No context found.

Gerhardt Brewka. Reasoning about priorities in default logic. In Barbara Hayes-Roth and Richard Korf, editors, Proceedings of the Twelfth National Conference on Arti- cial Intelligence, pages 940-945, Menlo Park, California, 1994. American Association for Articial Intelligence, AAAI Press. 16

First 50 documents  Next 50

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