Zobraziť v EDS

A Divide, Align and Conquer Strategy For Program Synthesis

Uložené v:
Podrobná bibliografia
Názov: A Divide, Align and Conquer Strategy For Program Synthesis
Autori: Witt, Jonas, Dumancic, Sebastijan, Guns, Tias, Carbon, Claus-Christian
Zdroj: Journal of Artificial Intelligence Research. 82:1961-1997
Informácie o vydavateľovi: AI Access Foundation, 2025.
Rok vydania: 2025
Predmety: Technology, 4603 Computer vision and multimedia computation, Science & Technology, 4611 Machine learning, 4602 Artificial intelligence, 0102 Applied Mathematics, 1702 Cognitive Sciences, Computer Science, 0801 Artificial Intelligence and Image Processing, Artificial Intelligence & Image Processing, Computer Science, Artificial Intelligence
Popis: A major bottleneck in search-based program synthesis, which learns programs from input/output examples, is the synthesis of large programs. As the size of the target program increases, so does the search depth, which leads to an exponentially growing number of candidate programs. Humans mitigate the combinatorial explosion that arises from deep program search: they build complex programs from smaller parts. We introduce a new strategy for program synthesis called Divide, Align & Conquer (DA&C) that exploits the compositionality of real-world domains to guide the synthesis towards useful subprograms. Divide decomposes each example using a segmentation procedure that is synthesized as part of the learning problem. Align matches the components in the decomposed input/output examples to steer the search toward combinations that lead to the synthesis of useful subprograms, and Conquer then solves a standalone synthesis problem on each pair of aligned input/output components. We show how replacing a deep program search with a linear number of much smaller synthesis tasks leads us to efficiently discover useful subprograms that are then combined into a solution program. Our agent outperforms current Inductive Logic Programming (ILP) methods on string transformation tasks even with minimal knowledge priors. Unlike existing methods, the predictive accuracy of our agent monotonically increases for additional examples. It approximates an average time complexity of O(m) in the size m of subprograms for highly structured and, hence, decomposable domains such as strings. Finally, we demonstrate the scalability of our technique on highdimensional abstract visual reasoning tasks from the Abstract Reasoning Corpus (ARC) for which ILP methods were previously infeasible. We are competitive with state-of-the-art agents outside of ILP, despite generating only 0.2% as many candidate programs from a knowledge prior of only 11 generic geometric primitives.
Druh dokumentu: Article
ISSN: 1076-9757
DOI: 10.1613/jair.1.16847
Prístupová URL adresa: https://lirias.kuleuven.be/handle/20.500.12942/765911
Rights: CC BY
Prístupové číslo: edsair.doi.dedup.....06ff2f4475f236bf15e39a66efa0fe88
Databáza: OpenAIRE
Popis
Abstrakt:A major bottleneck in search-based program synthesis, which learns programs from input/output examples, is the synthesis of large programs. As the size of the target program increases, so does the search depth, which leads to an exponentially growing number of candidate programs. Humans mitigate the combinatorial explosion that arises from deep program search: they build complex programs from smaller parts. We introduce a new strategy for program synthesis called Divide, Align & Conquer (DA&C) that exploits the compositionality of real-world domains to guide the synthesis towards useful subprograms. Divide decomposes each example using a segmentation procedure that is synthesized as part of the learning problem. Align matches the components in the decomposed input/output examples to steer the search toward combinations that lead to the synthesis of useful subprograms, and Conquer then solves a standalone synthesis problem on each pair of aligned input/output components. We show how replacing a deep program search with a linear number of much smaller synthesis tasks leads us to efficiently discover useful subprograms that are then combined into a solution program. Our agent outperforms current Inductive Logic Programming (ILP) methods on string transformation tasks even with minimal knowledge priors. Unlike existing methods, the predictive accuracy of our agent monotonically increases for additional examples. It approximates an average time complexity of O(m) in the size m of subprograms for highly structured and, hence, decomposable domains such as strings. Finally, we demonstrate the scalability of our technique on highdimensional abstract visual reasoning tasks from the Abstract Reasoning Corpus (ARC) for which ILP methods were previously infeasible. We are competitive with state-of-the-art agents outside of ILP, despite generating only 0.2% as many candidate programs from a knowledge prior of only 11 generic geometric primitives.
ISSN:10769757
DOI:10.1613/jair.1.16847