| Jorge, A., Brazdil, P. : Learning Recursion with Iterative Bootstrap Induction(extended abstract). In Proceedings of the European Conference on Machine Learning ECML-95 , Lavrac & Wrobel (Ed.), Springer-Verlag 1995
Brazdil, P., Jorge, A. : Learning by Refining Algorithm Sketches. In Proceedings of the Eleventh European Conference on Artificial Intelligence ECAI-94 , A. G. Cohn (Ed.),Wiley 1994
Jorge,A., Brazdil,P. "Architecture for Iterative Learning of Recursive Definitions" in L.De Raedt (ed.): Advances in Inductive Logic Programming, pp.206-218 IOS Press, Amsterdam 1996
|
| The system SkilIt described in [Jorge 95] is intended to induce recursive theories from small sets of positive examples using iterative bootstrapping. The system does not start from a scratch but it searches for a program meeting the input programming schema of the intended program. Moreover, SkilIt offers means to specify both
- some characteristics of the target predicate, e.g. its i/o mode, type declarations
- and the integrity constraints.
Auxiliary predicates are provided in the background theory (environment files). The present dataset includes training and testing data used for verification of functionality of SkilIt when inducing simple arithmetic and list processing concepts (e.g. sorting).
|