Authors: Céline Hocquette, Stephen Muggleton, Imperial College London
This repository contains the code for the paper Céline Hocquette and S.H. Muggleton. How much can experimental cost be reduced in active learning of agent strategies?. In Fabrizio Riguzzi, Elena Bellodi, and Riccardo Zese, editors, Proceedings of the 28th International Conference on Inductive Logic Programming, pages 38-53, Berlin, 2018. Springer-Verlag. available here.
These experiments require SWI-Prolog to run.
This work is based upon the Metagol system (Andrew Cropper and Stephen H. Muggleton, 2016, available at https://github.com/metagol/metagol) and the Metagol AI extension (Learning Higher-Order Logic Programs Through Abstraction and Invention, IJCAI 2016, Andrew Cropper and Stephen H. Muggleton)
for any query contact: celine.hocquette16@imperial.ac.uk