Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating Explanatory and Descriptive Learning in ILP. In Proceedings of the 15th International Joint Conference on Artificial Intelligence, IJCAI'97. Morgan Kaufman, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....rules are solved by additional priority relations. This framework captures the notion of speci city of a rule as it is done in [3] in prioritized default logic. But, it is known [23] that speci city can be handled by means of semi normal defaults and that is exactly what our method does. In [7] the problem of contradiction between de nition of p and :p is solved by using integrity constraints in order to restrict the conclusions derivable from too general rules. More recently, some works deal with this problem in the context of extended logic programs [11, 13, 12] Extended Logic ....

Y. Dimopoulos, S. Deroski, and A. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the International Joint Conference on Articial Intelligence, volume 2, pages 900906. Morgan Kaufmann Publishers, 1997.

....to specialize overgeneral e ect axioms instead of fully specialize them by adding preconditions. The choice of the constraints is not independent from the rules of the theory, and the diculty lies in nding the relevant constraints that would compensate correctly for the rules of the theory [6]. 6 Conclusions A logic programming formalization of action domains has been well studied, yet not much previous work [21] exists on the combination with learning methods. The presented approach improves on previous ones by providing more expressivity, so that it can solve some problems that ....

Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the 15th International Joint Conference on Arti cial Intelligence (IJCAI-97), pages 900-907, San Francisco, 1997. Morgan Kaufmann Publishers.

....the select rule and the constraint learned separately do not achieve the same e ect. Furthermore, the above constraint can exist independently of any negative examples. As a counterpart, the diculty is to nd the relevant constraints that would compensate correctly for the rules of the theory [3]. This is part of our current work. Let us consider the Logistics domain [15] In this domain we have two types of vehicles: trucks and airplanes. Trucks can be used to transport goods within a city, and airplanes can be used to transport goods between two airports. The problems in this domain ....

Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the 15th International Joint Conference on Articial Intelligence (IJCAI-97), pages 900-907, San Francisco, 1997. Morgan Kaufmann Publishers.

....predicate learning in ILP because there are no proofs involved, and so mutual recursion is not a problem. Finally, there is the task of learning mixed theories of predicate definitions and integrity constraints. Previous work has concentrated on learning them separately by two different learners [6]. Tertius is able to learn such mixed theories in one go. 7. Conclusions In this paper we described the Tertius approach to first order unsupervised learning. First order logic offers the ability to deal with structured, multi relational knowledge. Tertius is a general purpose first order ....

Y. Dimopoulos, S. Dzeroski, and A.C. Kakas. Integrating explanatory and descriptive learning in ILP. In M.E. Pollack, editor, Proceedings of the 15th International Joint Conference on Artificial Intelligence, pages 900--906, 1997.

....formalism available for specifying background knowledge in the Tertius first order logic discovery tool. It also describes the way background knowledge effects the results of Tertius. The way Tertius handles integrity constraints is compared to an approach suggested in work done on the CLAUDIEN [1] learning system. Finally a number of examples of use of Tertius for analysing the audiology dataset 1 are also provided. 1 The audiology data set is taken from the University of California at Irvine s Machine Learning Repository; http: www.ics.uci.edu AI ML MLDBRepository.html Contents ....

....is an improvement on other learning systems because it can use integrity constraints in its background knowledge. In particular it compares using Tertius with using CLAUDIEN in a integrity constraint setting that was originally presented to show how CLAUDIEN can handle integrity constraints [1]. Combining explanatory and descriptive Inductive Logic Programming can be done in a framework of declarative background knowledge with integrity constraints [1] This framework is more expressive than normal horn clauses with negation as failure [2] and allows the learning of inherent ....

[Article contains additional citation context not shown here]

Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating Explanatory and Descriptive Learning in ILP. In Proceedings of the 15th International Joint Conference on Artificial Intelligence, IJCAI'97. Morgan Kaufman, 1997.

....predicate learning in ILP because there are no proofs involved, and so mutual recursion is not a problem. Finally, there is the task of learning mixed theories of predicate definitions and integrity constraints. Previous work has concentrated on learning them separately by two different learners [3]. Tertius is able to learn such mixed theories in one go. 1.3. Outline of the paper The outline of the paper is as follows. In Section 2 we present the preliminary material necessary to understand the rest of the paper, in particular regarding our learning model of confirmatory induction, and ....

Yannis Dimopoulos, Saso Dzeroski, and Antonis Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence (IJCAI'97), pages 900--906, Nagoya, Japan, 1997.

No context found.

) Y. Dimopoulos, S. Dzeroski, and A. C. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the 15th International Joint Conference on Artificial Intelligence, 1997.

....to learn is itself inherently abductive or non monotonic (e.g. containing nested hierarchies of exceptions) in which case the hypothesis space for learning is a space of abductive theories. Frameworks for learning abductive theories include the early system of LAB [14] and the more recent work of [7, 8, 3, 9] where both explanatory and confirmatory induction are used to generate abductive theories that include integrity constraints. Also in [6] the problem of learning abductive logic programs for capturing non monotonic theories is studied. 5 Concluding Remarks The cycle of cooperation between ....

Yannis Dimopoulos, Saso Dzeroski, and Antonis C. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence, pages 900--905. Morgan Kaufmann, 1997.

....been the topic of study of an ECAI96 workshop [MRFK96] Our work builds on earlier work in [DK96] and [ELM 96, LMMR97] for learning simpler forms of abductive theories. Finally, the issue of integrating discriminant (or explanatory) and characterizing ILP systems has also been put forward in [DDK96] Recently, there have been several other proposals for learning with incomplete information. The FOIL I system [IKI 96] learns from incomplete information in the training examples set but not in the background knowledge, with a particular emphasis on learning recursive predicates. Similarly ....

Y. Dimopoulos, S. Dzeroski, and A.C. Kakas. Integrating explanatory and descriptive learning in ilp. Technical Report TR-96-16, University of Cyprus, Computer Science Department, 1996.

No context found.

Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating Explanatory and Descriptive Learning in ILP. In Proceedings of the 15th International Joint Conference on Artificial Intelligence, IJCAI'97. Morgan Kaufman, 1997.

No context found.

Y. Dimopoulos, S. Dzeroski, and A. Kakas. Integrating explanatory and descriptive learning in ILP. In Proceedings of IJCAI97, pages 900--906, 1997.

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