Imperial College London
Portrait Picture

ProfessorStephenMuggleton

Faculty of Engineering, Department of Computing

Royal Academy Chair in Machine Learning
 
 
 
//

Contact

 

+44 (0)20 7594 8307s.muggleton Website

 
 
//

Assistant

 

Mrs Bridget Gundry +44 (0)20 7594 1245

 
//

Location

 

407Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Cropper:2020:10.1007/s10994-019-05862-7,
author = {Cropper, A and Morel, R and Muggleton, S},
doi = {10.1007/s10994-019-05862-7},
journal = {Machine Learning},
pages = {1289--1322},
title = {Learning higher-order logic programs},
url = {http://dx.doi.org/10.1007/s10994-019-05862-7},
volume = {109},
year = {2020}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - A key feature of inductive logic programming is its ability to learn first-order programs, which are intrinsically more expressive than propositional programs. In this paper, we introduce techniques to learn higher-order programs. Specifically, we extend meta-interpretive learning (MIL) to support learning higher-order programs by allowing for higher-order definitions to be used as background knowledge. Our theoretical results show that learning higher-order programs, rather than first-order programs, can reduce the textual complexity required to express programs, which in turn reduces the size of the hypothesis space and sample complexity. We implement our idea in two new MIL systems: the Prolog system Metagol ho and the ASP system HEXMIL ho. Both systems support learning higher-order programs and higher-order predicate invention, such as inventing functions for map/3 and conditions for filter/3. We conduct experiments on four domains (robot strategies, chess playing, list transformations, and string decryption) that compare learning first-order and higher-order programs. Our experimental results support our theoretical claims and show that, compared to learning first-order programs, learning higher-order programs can significantly improve predictive accuracies and reduce learning times.
AU  - Cropper,A
AU  - Morel,R
AU  - Muggleton,S
DO  - 10.1007/s10994-019-05862-7
EP  - 1322
PY  - 2020///
SN  - 0885-6125
SP  - 1289
TI  - Learning higher-order logic programs
T2  - Machine Learning
UR  - http://dx.doi.org/10.1007/s10994-019-05862-7
VL  - 109
ER  -
Download