Talks and Poster Presentations (without Proceedings-Entry):

T. Kaminski, T. Eiter, K. Inoue:
"Efficiently Encoding Meta-Interpretive Learning by Answer Set Programming";
Talk: 28th International Conference on Inductive Logic Programming, Ferrara, Italy; 2018-09-02 - 2018-09-04.

English abstract:
Meta-Interpretive Learning (MIL) learns logic programs from examples by instantiating meta-rules, which is implemented by the Metagol system using Prolog. Based on previous work wrt. solving MIL by employing Answer Set Programming, in this work-in-progress paper we describe a modification of a previous MIL-encoding which prunes the search space more effectively by sim ulating a top-down search as performed by Prolog. Initial experiments show that our new encoding can significantly speed up the inductive learning process.

Electronic version of the publication:

Related Projects:
Project Head Thomas Eiter:
Integrated Evaluation of Answer Set Programs and Extensions

Created from the Publication Database of the Vienna University of Technology.