Daniel Borrajo, Juan P. Caraqa-Valente, and Juan Pazos. A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland, 1992.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....rules. The preconditions are always sorted according to an ordering function that prefers the preconditions that: ffl refer to more variables that already appear in preconditions of the control rule in this way, we keep the locality heuristic that was proposed in earlier work (Muggleton, 1992, Borrajo et al. 1992a, Borrajo et al. 1992b, Borrajo and Veloso, 1994) and that empirical results show was effective ; 8 hamlet creates this pool from the rest of preconditions that were not introduced directly into the rule when it was created. ffl refer to literals that have appeared more times in the ....

.... are always sorted according to an ordering function that prefers the preconditions that: ffl refer to more variables that already appear in preconditions of the control rule in this way, we keep the locality heuristic that was proposed in earlier work (Muggleton, 1992, Borrajo et al. 1992a, Borrajo et al. 1992b, Borrajo and Veloso, 1994) and that empirical results show was effective ; 8 hamlet creates this pool from the rest of preconditions that were not introduced directly into the rule when it was created. ffl refer to literals that have appeared more times in the preconditions of control rules ....

Daniel Borrajo, Juan P. Cara¸ca-Valente, and Juan Pazos. A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland, 1992.

....rules. The preconditions are always sorted according to an ordering function that prefers the preconditions that: refer to more variables that already appear in preconditions of the control rule in this way, we keep the locality heuristic that was proposed in earlier work (Muggleton, 1992, Borrajo et al. 1992a, Borrajo et al. 1992b, Borrajo and Veloso, 1994) and that empirical results show was effective ; SHAMLET creates this pool from the rest of preconditions that were not introduced directly into the rule when it was created. refer to literals that have appeared more times in the preconditions ....

.... are always sorted according to an ordering function that prefers the preconditions that: refer to more variables that already appear in preconditions of the control rule in this way, we keep the locality heuristic that was proposed in earlier work (Muggleton, 1992, Borrajo et al. 1992a, Borrajo et al. 1992b, Borrajo and Veloso, 1994) and that empirical results show was effective ; SHAMLET creates this pool from the rest of preconditions that were not introduced directly into the rule when it was created. refer to literals that have appeared more times in the preconditions of control rules of the ....

Daniel Borrajo, Juan P. Caraqa-Valente, and Juan Pazos. A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland, 1992.

....under which the decision was made and also define the situations under which the rule can be reapplied. The appropriate set of features that we consider in our bounded explanation technique as well as the credit assignment procedure have evolved from extensive previous work of the first author [ Borrajo et al. 1992b ] Although there is no guarantee that this set of features is a sufficient set, there have been a number of iterations in the design of the set, to generate our confidence on it. Furthermore the empirical experiments confirm that the set is appropriate and the induction (refinement method) ....

....be unified by two variables that belong to classes that are subclasses of a common class (except for the root) this operator generalizes the variables to the common superclass. We implemented previously a variation of this technique applied to a the parametrization procedure of a single rule [ Borrajo et al. 1992a ] This set of inductive operators may produce overgeneral rules in special situations, as we have been experiencing in our more sophisticated recent tests. These situations are beneficial for our inductive learning style as they provide negative examples of the application of the learned ....

Daniel Borrajo, Juan P. Caraca-Valente, and Juan Pazos. A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland, 1992.

....control rule need to establish the relevant conditions under which the decision was made and also define the situations under which the rule can be re applied. The appropriate set of features that we consider in our bounded explanation technique has evolved from previous work of the first author [3]. Although there is no guarantee that this set of features is a sufficient set, there have been a number of iterations in the design of the set, to generate our confidence on it. Furthermore the empirical experiments confirm that the set is appropriate and the induction and refinement phases ....

Daniel Borrajo, Juan P. Caraca-Valente, and Juan Pazos. A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland, 1992.

No context found.

Borrajo, D., Cara¸ca-Valente, J. P., and Pazos, J. (1992b). A knowledge compilation model for learning heuristics. In Proceedings of the ML92 Workshop on Knowledge Compilation and Speedup Learning, Aberdeen, Scotland.

....or over general. Upon experiencing new problem solving episodes, HAMLET refines its control knowledge, incrementally acquiring increasingly correct control knowledge (Borrajo and Veloso, 1993; Borrajo and Veloso, 1994b) HAMLET evolves from previous work from one of the authors in NOLIMIT (Borrajo et al. 1992). The inputs to HAMLET are a domain, a set of training problems, and a quality measure. 4 The output is a set of control rules. HAMLET has three main modules: the Bounded Explanation module, the Inductive module, and the Refinement module. The Bounded Explanation module generates control rules ....

Borrajo, D., Caraca-Valente, J. P., and Pazos, J. (1992). A knowledge compilation model for learning heuristics. In Proceedings of the Workshop on Knowledge Compilation of the 9th International Conference on Machine Learning, Scotland.

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