Optimal Learning of Specifications from Examples

TL;DR

This paper introduces SPEX, an efficient learning algorithm that minimizes user questions while accurately identifying hypotheses in domains like stock trading and data structures, ensuring practical usability.

Contribution

SPEX is a novel algorithm that guarantees minimal questions in learning formulas over first-order predicates using membership queries, even with correlated predicates.

Findings

SPEX drastically reduces the number of questions needed.

It effectively learns hypotheses in diverse domains.

Experimental results confirm its practical efficiency.

Abstract

A fundamental challenge in synthesis from examples is designing a learning algorithm that poses the minimal number of questions to an end user while guaranteeing that the target hypothesis is discovered. Such guarantees are practically important because they ensure that end users will not be overburdened with unnecessary questions. We present SPEX -- a learning algorithm that addresses the above challenge. SPEX considers the hypothesis space of formulas over first-order predicates and learns the correct hypothesis by only asking the user simple membership queries for concrete examples. Thus, SPEX is directly applicable to any learning problem that fits its hypothesis space and uses membership queries. SPEX works by iteratively eliminating candidate hypotheses from the space until converging to the target hypothesis. The main idea is to use the implication order between hypotheses to guarantee that in each step the question presented to the user obtains maximal pruning of the space. This problem is particularly challenging when predicates are potentially correlated. To show that SPEX is practically useful, we expressed two rather different applications domains in its framework: learning programs for the domain of technical analysts (stock trading) and learning data structure specifications. The experimental results show that SPEX's optimality guarantee is effective: it drastically reduces the number of questions posed to the user while successfully learning the exact hypothesis.