mlsanityData to 14 Oct 2026

Learning logic programs by discovering where not to search

arXiv 2202.09806 ↗First posted 20 Feb 2022cs.LG cs.AI cs.LOcited by 4

Not on the map. Why?

Abstract (arXiv)

The goal of inductive logic programming (ILP) is to search for a hypothesis that generalises training examples and background knowledge (BK). To improve performance, we introduce an approach that, before searching for a hypothesis, first discovers where not to search. We use given BK to discover constraints on hypotheses, such as that a number cannot be both even and odd. We use the constraints to bootstrap a constraint-driven ILP system. Our experiments on multiple domains (including program synthesis and game playing) show that our approach can (i) substantially reduce learning times by up to 97%, and (ii) scale to domains with millions of facts.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Builds on 11 of its 61 references that are in the map. First posted on arXiv 20 Feb 2022.

Builds on

11 of 61 references.

References · 11

  1. 2017Learning Explanatory Rules from Noisy Datacited by 211
  2. 2020Inductive logic programming at 30: a new introductioncited by 85
  3. 2020Learning programs by learning from failurescited by 73
  4. 2021Inductive logic programming at 30cited by 47
  5. 2019Making sense of sensory inputcited by 29
  6. 2020Randomised Gaussian Process Upper Confidence Bound for Bayesian Optimisationcited by 22
  7. 2019Logical reduction of metarulescited by 19
  8. 2019Inductive general game playingcited by 18
  9. 2019Learning Relational Representations with Auto-encoding Logic Programscited by 13
  10. 2019Towards meta-interpretive learning of programming language semanticscited by 5
  11. 2021Learning logic programs through divide, constrain, and conquercited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 1.

4 papers.

  1. 2025Learning Logical Rules using Minimum Message Lengthcited by 1
  2. 2025Honey, I shrunk the hypothesis space (through logical preprocessing)cited by 0
  3. 2025Symmetry breaking for inductive logic programmingcited by 0
  4. 2026HyDI: A hybrid Deep Learning-Inductive Logic Programming ensemble for multi-label classificationcited by 0