Learning Higher-Order Programs without Meta-Interpretive Learning

TL;DR

This paper introduces a novel extension to the Learning From Failures paradigm that leverages higher-order definitions, significantly enhancing the efficiency and effectiveness of learning complex programs in inductive logic programming without extensive human guidance.

Contribution

It presents a new theoretical framework and practical extension for higher-order ILP that improves learning performance while maintaining soundness and reducing human intervention.

Findings

Enhanced learning accuracy and performance in ILP tasks

Significant reduction in human guidance needed

Theoretical validation of higher-order definition class

Abstract

Learning complex programs through inductive logic programming (ILP) remains a formidable challenge. Existing higher-order enabled ILP systems show improved accuracy and learning performance, though remain hampered by the limitations of the underlying learning mechanism. Experimental results show that our extension of the versatile Learning From Failures paradigm by higher-order definitions significantly improves learning performance without the burdensome human guidance required by existing systems. Our theoretical framework captures a class of higher-order definitions preserving soundness of existing subsumption-based pruning methods.