AbstractBeam: Enhancing Bottom-Up Program Synthesis using Library Learning

TL;DR

AbstractBeam improves program synthesis by learning and leveraging recurring program structures, significantly outperforming LambdaBeam in efficiency and success rate within list manipulation tasks.

Contribution

The paper introduces AbstractBeam, a framework that enhances LambdaBeam with Library Learning to incorporate recurring program blocks, improving synthesis performance.

Findings

AbstractBeam outperforms LambdaBeam in task success rate.

AbstractBeam requires less time and fewer candidates to synthesize programs.

Library Learning broadens applicability beyond specific domains.

Abstract

LambdaBeam is a state-of-the-art, execution-guided algorithm for program synthesis that utilizes higher-order functions, lambda functions, and iterative loops within a Domain-Specific Language (DSL). LambdaBeam generates each program from scratch but does not take advantage of the frequent recurrence of program blocks or subprograms commonly found in specific domains, such as loops for list traversal. To address this limitation, we introduce AbstractBeam: a novel program synthesis framework designed to enhance LambdaBeam by leveraging Library Learning. AbstractBeam identifies and integrates recurring program structures into the DSL, optimizing the synthesis process. Our experimental evaluations demonstrate that AbstractBeam statistically significantly (p < 0.05) outperforms LambdaBeam in the integer list manipulation domain. Beyond solving more tasks, AbstractBeam's program synthesis is also more efficient, requiring less time and fewer candidate programs to generate a solution. Furthermore, our findings indicate that Library Learning effectively enhances program synthesis in domains that are not explicitly designed to showcase its advantages, thereby highlighting the broader applicability of Library Learning.