Abstracting Concepts with Inverse Resolution
Abstract
Usually, abstraction is defined as a mapping between languages. In this paper it is, instead, defined as a mapping between models and extends Tenenberg's restricted predicate mapping. The mapping is axiomatized by means of a theory TA defining the semantics of the relations in the abstract model from those existing in the ground model. Therefore, this form of abstraction is semantic and must be evaluated using a deductive mechanism. A restricted class of semantic abstraction (CP-abstraction) is characterized, having the property of preserving the concept instances with respect to a given model. CP-abstraction fits into the paradigm of inverse resolution, which has been proposed as a framework for constructive learning. A restricted form of absorption rule is also introduced to compute it. Finally, a new operation, called term abstraction, is defined; it allows abstraction mechanisms, similar to the ones used to handle abstract data types, to be implemented.
Cite
Text
Giordana et al. "Abstracting Concepts with Inverse Resolution." International Conference on Machine Learning, 1991. doi:10.1016/B978-1-55860-200-7.50032-5Markdown
[Giordana et al. "Abstracting Concepts with Inverse Resolution." International Conference on Machine Learning, 1991.](https://mlanthology.org/icml/1991/giordana1991icml-abstracting/) doi:10.1016/B978-1-55860-200-7.50032-5BibTeX
@inproceedings{giordana1991icml-abstracting,
title = {{Abstracting Concepts with Inverse Resolution}},
author = {Giordana, Attilio and Saitta, Lorenza and Roverso, Davide},
booktitle = {International Conference on Machine Learning},
year = {1991},
pages = {142-146},
doi = {10.1016/B978-1-55860-200-7.50032-5},
url = {https://mlanthology.org/icml/1991/giordana1991icml-abstracting/}
}