S. A. Goldman, S. S. Kwek, and S. D. Scott. Learning from examples with unspecified attribute values. In Proceedings of the 10th Annual Conference on Computational Learning Theory, pages 231--242. ACM Press, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....that gives a unique characterization for any set of example based queries. However, there are queries that do not fit in the notion of example based queries, for instance restricted equivalence [1] equivalence queries that do not supply counterexample) and unspecified attribute value queries [9] (UAVs) There are works on lower bounds for UAVs [5, 12] but, as far as we know, there is no general learning dimension for UAVs comparable to AIdim in the case of example based queries. In this work we introduce a new combinatorial notion, Gdim, that gives a good approximation of the number of ....

....(more than a constant) to discover a cheating teacher and therefore we cannot rule out the answering schemes that are not satisfiable. 5 5 Unspecified Attribute Value protocols In this section we present our results for a concrete query set: the Unspecified Attribute Value (UAV) queries from [9, 5, 6]. Following [9] given a function f 2 B n and a partial assignment ff 2 f0; 1; g n , we define f(ff) 1 if and only if f is the constant 1 in H ff . Similarly, we define f(ff) 0 if and only if f is the constant 0 in H ff . Otherwise, we say that f(ff) for unknown) For a target concept ....

[Article contains additional citation context not shown here]

S. A. Goldman, S. S. Kwek, and S. D. Scott. Learning from examples with unspecified attribute values. In Proc. 10th Annu. Conf. on Comput. Learning Theory, pages 231-- 242. ACM Press, New York, NY, 1997.

....Attribute Values Andreas Birkendorf, Norbert Klasner, Christian Kuhlmann, and Hans U. Simon Lehrstuhl Mathematik und Informatik Fakultat fur Mathematik Ruhr Universitat Bochum D 44780 Bochum February 9, 1998 Abstract This paper deals with the UAV learning model of Goldman, Kwek and Scott [7]. UAV is the acronym for Unspecified Attribute Values . As in [7] we consider exact learning within the UAV framework, where the learner has to exactly identify an unknown target concept by means of UAV membership (UAV MQs) and or UAV equivalence queries (UAV EQs or UAV ARB EQs, ....

....and Hans U. Simon Lehrstuhl Mathematik und Informatik Fakultat fur Mathematik Ruhr Universitat Bochum D 44780 Bochum February 9, 1998 Abstract This paper deals with the UAV learning model of Goldman, Kwek and Scott [7] UAV is the acronym for Unspecified Attribute Values . As in [7], we consider exact learning within the UAV framework, where the learner has to exactly identify an unknown target concept by means of UAV membership (UAV MQs) and or UAV equivalence queries (UAV EQs or UAV ARB EQs, respectively) A smooth transition between exact learning in the UAV setting ....

[Article contains additional citation context not shown here]

Sally A. Goldman, Stephen Kwek, and Stephen D. Scott. Learning from examples with unspecified attribute values. In Proceedings of the 10th Annual Conference on Computational Learning Theory, pages 231--242. ACM Press, New York, NY, 1997.

....by ESPRIT Project ALCOM IT (Project 20244) PAC learning theory the situation is called Restricted Focus of Attention (RFA) learning, introduced in [3, 4, 6] see [15] for an extensive survey. For query based learning the associated framework is the Unspecified Attribute Values learning of [17]. A good example of a data set that motivates the work here is a medical prognosis problem analysed in Titterington et al. 26] and Lowe and Webb [23] The data represent 1000 head injured coma patients, and contains (for each patient) a subset of a set of 6 diagnostic indicators measured on ....

S.A. Goldman, S.S. Kwek and S.D. Scott (1997). Learning from Examples with Unspecified Attribute Values. Tenth annual COLT conference, 231-242, ACM Press, New York.

.... Computer Science University of Calgary 2500 University Drive NW Calgary, AB, Canada T2N 1N4 Email: fbshouty, wilsondg cpsc.ucalgary.ca Abstract We continue the study of learning in the presence of unspecified attribute values (UAV) where some of the attributes of the examples may be unspecified [9, 4]. A UAV assignment x 2 f0; 1; g n , where indicates unspecified, is classified positive (negative) with respect to a Boolean function f if all possible assignments for the unspecified attributes result in a positive (negative) classification. Otherwise, the classification of x is . Given an ....

....trees and the class of Boolean functions of a constant number of terms or clauses. The former of the two previous results leads to a quasi polynomial time UAV MQ algorithm for decision trees with polynomial size and CDNF with polynomial size. Finally, we answer an open problem posed in both [9] and [4] by showing that decision trees are learnable using UAV MQs and UAV EQs. 1 Introduction In an effort to better understand situations that are subjected to incomplete information, Goldman, Kwek and Scott [9] have introduced a situation in which the Boolean examples may have unspecified ....

[Article contains additional citation context not shown here]

Sally A. Goldman, Steven Kwek, and Stephen D. Scott. Learning from examples with unspecified attribute values. In Proceedings of the Tenth Annual Conference on Computational Learning Theory, pages 231--242, 1997.

....as would be needed to use the algorithm of Bshouty et. al to learn DNF formulas with only a UAV MQ oracle. Recently, Birkendorf, Klasner, Kuhlman and Simon [11] investigated the UAV model further and answered a number of open problems posted in an earlier (conference) version of this paper [24]. They presented lower bound results on the number of UAV EQs and UAV MQs required to learn a concept class in terms of its Vapnik Chervonenkis dimension. Further, they extended Angluin s [3] sunflower lemma (which is useful in proving lower bound results in the exact model) to the UAV setting. In ....

Sally A. Goldman, Stephen S. Kwek, and Stephen D. Scott. Learning from examples with unspecified attribute values. In Proc. 10th Annu. Conf. on Comput. Learning Theory, pages 231--242. ACM Press, New York, NY, 1997.

....as would be needed to use the algorithm of Bshouty et. al to learn DNF formulas with only a UAV MQ oracle. Recently, Birkendorf, Klasner, Kuhlman and Simon [11] investigated the UAV model further and answered a number of open problems posted in an earlier (conference) version of this paper [24]. They presented lower bound results on the number of UAV EQs and UAV MQs required to learn a concept class in terms of its Vapnik Chervonenkis dimension. Further, they extended Angluin s [3] sunflower lemma (which is useful in proving lower bound results in the exact model) to the UAV setting. In ....

Sally A. Goldman, Stephen S. Kwek, and Stephen D. Scott. Learning from examples with unspecified attribute values. In Proc. 10th Annu. Conf. on Comput. Learning Theory, pages 231--242. ACM Press, New York, NY, 1997.

No context found.

S. A. Goldman, S. S. Kwek, and S. D. Scott. Learning from examples with unspecified attribute values. In Proceedings of the 10th Annual Conference on Computational Learning Theory, pages 231--242. ACM Press, 1997.

No context found.

S.A. Goldman, S. S. Kwek adn S. D. Scott, Learning from examples with unspecified attribute values, Proceedings of COLT'97 (1997) 231-242.

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