In plain words: Inductive logic programming learns logical rules that fit training examples and generalize to new ones. A review of the past decade tracks progress in searching rule space, learning self-calling rules, inventing new concepts, and mixing in other technologies, then lists what holds it back.

Abstract · Inductive logic programming at 30

Inductive logic programming (ILP) is a form of logic-based machine learning. The goal is to induce a hypothesis (a logic program) that generalises given training examples. As ILP turns 30, we review the last decade of research. We focus on (i) new meta-level search methods, (ii) techniques for learning recursive programs, (iii) new approaches for predicate invention, and (iv) the use of different technologies. We conclude by discussing current limitations of ILP and directions for future research.

Andrew Cropper, Sebastijan Dumančić, Richard Evans, Stephen H. Muggleton
arXiv:2102.10556 · cs.AI, cs.LG · submitted Feb 21, 2021 · updated Sep 22, 2021
abstract · pdf · html · Extension of IJCAI20 survey paper. Accepted for the MLJ. arXiv admin note: substantial text overlap with arXiv:2002.11002, arXiv:2008.07912