Alon Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 588--594, 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....MI abstractions have the advantage over the syntactic ones in [Giunchiglia and Walsh, 1992] in that they capture more of the underlying justification that leads to the abstraction, Nayak and Levy, 1995] Among other insights, the work shows how the ABSTRIPS abstraction is an MI one. [Levy, 1994] formalizes irrelevance of clauses with respect to queries on knowledge bases, as well as independence of predicate arguments. Nayak, 1994] combines abstractions within the theory of contexts. 5 Rich and Poor Objects A rich object is one that cannot be completely described, while a poor one ....

Levy, A. Y. (1994). Creating abstractions using relevance reasoning. In AAAI, Vol. 1, pages 588--594.

....sequent calculus) thus the problem space was not abstracted at all. In contrary Knoblock s algorithm, we use here, produces reduced models, where literals are abstracted away at particular abstraction level from all formulae. Other approaches to automatic generation of abstraction spaces include [2, 18, 63]. We are exploiting abstraction as a mechanism to reduce number of literals used during inference. An abstraction level determines which literals can be used during inference. Either in terms of Petri net markings or LL progam context. If an abstract space is formed by dropping conditions ....

A. Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Arti cial Intelligence (AAAI'94), pp. 588-594, AAAI Press/The MIT Press, 1994.

....in preconditions, thus the problem space was not abstracted at all. In contrary Knoblock s algorithm, we use here, produces reduced models, where literals are abstracted away at particular abstraction level from all formulae. Other approaches to automatic generation of abstraction spaces include [1, 2, 11]. In this paper we presented a way to apply abstractions to LL sequents using information about dependencies between literals in LL extralogical axioms. Then through hierarchical theorem proving the complexity of overall theorem proving would be decreased. If an abstract space is formed by ....

A. Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Arti cial Intelligence (AAAI'94), pages 588-594, 1994.

....as in Wiegel and Bliek s work or similar polynomial time simplification techniques can be used in the CSPs themselves. All of these techniques require explicit constraints. Levy s abstraction by relevance reasoning system abstracts a knowledge base to fit a specific query or set of queries [ Levy, 1994 ] The system creates an abstraction with both the downward and upward solution properties and which is guaranteed to be at least as e#cient, computationally, as the original knowledge base. Abstraction according to relevance might be applicable to a compiled planning problem with the action and ....

Alon Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence. Menlo Park, Calif.: AAAI Press, July 1994.

....in a straightforward manner. 4 Discussion 4.1 Impasse Driven Abstraction There has been much recent research towards the automatic generation of abstractions. Such work has primarily been focused on developing abstractions by performing pre task analyses of a problem solving domain, e.g. [4, 9, 12, 15, 23]. The approach described here does not rely on precomputed analyses, but draws on different sources of knowledge, derived from run time context, to determine the abstractions. Because Spatula s knowledge is obtained during problem solving, it can be applied when pre task analysis does not produce ....

A. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 588--594. AAAI Press/The MIT Press, 1994.

....I) The purpose of both variable elimination and the operator splitting technique is to obtain a smallsized theory upon instantiation. Although we think that we discussed the major techniques for equivalent KB reformation, other methods might be of interest as well. For instance, Levy [12] proposes theory abstraction as a method to transform a given 23 theory to a (computationally) simpler one. The idea of theory abstraction is to remove some detail from the knowledge base, in particular, some argument of a predicate, if the argument is irrelevant to answering a given query type. ....

Alon Y. Levy. Creating abstractions using relevance reasoning. In Proceedings 12th National Conference on Artificial Intelligence (AAAI94) , pages 588--594, 1994.

....should then be used to hide unimportant information and only summarize in a pertinent manner the data he must see. Abstraction were for example used in theorem proving [13] planning [20,19] first order logic [17] learning [27] digital circuits troubleshooting [14] and knowledge bases [21]. Abstraction and reformulation techniques have already been applied to permit more efficient solution of a CSP. Choueiry and Faltings [4] relate interchangeability to abstraction in the context of a decomposition heuristic for resource allocation. Weigel and Faltings [32] cluster variables to ....

A. Y. Levy, Creating Abstractions Using Relevance Reasoning, in: Proc. of the Workshop on Theory Reformulation and Abstraction, pp. 2-139/2-153, 1994.

No context found.

Alon Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 588--594, 1994.

....underlying modeling assumptions being made, i.e. on the abstractions of the domain and the approximations being made. Deciding to make a certain abstraction can be viewed as stating that some detail is irrelevant to the goal and can hence be removed from the representation of a phenomenon (see [ Levy, 1994 ] for a more detailed account of the connection between irrelevance and abstractions) We therefore use relevance reasoning to decide which modeling assumptions are appropriate for the goal. Furthermore, explicit relevance claims can be used to express additional domain knowledge that comes to ....

....in the chosen scenario model may be satisfied at a certain step of the simulation and cease to hold at a later point, modeling assumptions are assumed to hold throughout the simulation. We make the following assumptions about the library of model fragments. Their formal definitions are given in [ Levy et al. 1994 ] Coherence of the Library: The library coherence assumption requires that if we have a set of model fragments that have consistent modeling assumptions and whose operating conditions are satisfied, then the resulting set of equations will not be over constrained (i.e. will not have more ....

[Article contains additional citation context not shown here]

Levy, Alon Y. 1994. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence.

....that needs to be explored is unnecessarily large. Often, it is possible to detect efficiently that facts (or sets of facts) in the knowledge base are irrelevant to a query [30, 21, 16] or to detect that a knowledge base can be abstracted and still be able to answer a set of queries correctly [31, 17]. In other cases, it is possible to give the system additional meta level control advice as to which facts in the knowledge base are possibly relevant to a query, therefore enabling the inference mechanism to ignore the rest. Levy has developed a general framework for reasoning about relevance ....

....to build models (e.g. 25, 33, 27] 4.2.2 Selecting the Level of Detail The second part of the model selection problem is determining the level of detail at which to model each phenomenon. This entails deciding which abstractions and approximations can be made in modeling the system. Levy in [17, 16] demonstrates that knowledge underlying such decisions can be stated as relevance claims and better understood when stated in that form. In our algorithm, we bring relevance knowledge to bear in choosing the level of detail in two ways: ffl We articulate the difference between CMFs in an ....

Alon Y. Levy. Creating abstractions using relevance reasoning. In Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 588--594, Menlo Park, CA, 1994. AAAI Press/The MIT Press.

....hence (M base ) is a model of both T abs and S abs , and hence a model of T abs [ S abs . 2 Compositionality is exploited in diagnosis with multiple theories [ Nayak, 1994b ] and in compositional modeling [ Falkenhainer and Forbus, 1991; Nayak, 1994a; Nayak and Joskowicz, 1996; Iwasaki and Levy, 1994 ] where theories are built by composing knowledge from different sources. Moreover, one may argue that independent of these applications, theory compositionality is intrinsic to the very notion of abstractions. When T abs is an MI abstraction of T base , T abs can have models that are not the ....

....of a set of simplifying assumptions and an MI abstraction has two key advantages. First, the simplifying assumptions underlying the abstraction are made explicit, and therefore can be used in reasoning, as has been done in compositional modeling [ Falkenhainer and Forbus, 1991; Iwasaki and Levy, 1994 ] and diagnosis [ Davis, 1984; Nayak, 1994b; Struss, 1992 ] Second, we can show that an abstraction will yield false proofs only if the simplifying assumptions are inappropriate. In particular, a simple corollary of Proposition 1 is that if an abstraction of a consistent base theory is ....

[Article contains additional citation context not shown here]

Levy, A. Y. 1994. Creating abstractions using relevance reasoning. In Procs. of AAAI-94.

....of ignoring irrelevant information, and on problems that seemed previously unrelated. A key component of our work addressed the issue of developing efficient algorithms for automatically detecting irrelevant parts of a knowledge base [ Levy and Sagiv, 1992; Levy et al. 1993; Levy and Sagiv, 1993; Levy et al. 1994a ] Our work yielded solutions to several open theoretical problems, as well as practical algorithms which are now being incorporated into commercial database systems. We have also applied our framework to the problems of automatically creating abstractions [ Levy, 1994 ] creating models for ....

....and Sagiv, 1993; Levy et al. 1994a ] Our work yielded solutions to several open theoretical problems, as well as practical algorithms which are now being incorporated into commercial database systems. We have also applied our framework to the problems of automatically creating abstractions [ Levy, 1994 ] creating models for physical devices (for tasks such as design, simulation and diagnosis) Iwasaki and Levy, 1994 ] and gathering information in distributed heterogeneous environments [ Levy et al. 1994b ] This paper focuses on the foundations of our framework and outlines the space of ....

[Article contains additional citation context not shown here]

Levy, Alon Y. 1994. Creating abstractions using relevance reasoning. In Proceedings of AAAI-94.

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