Machine Learning for Quantifier Selection in cvc5

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jakubův, Jan, Janota, Mikoláš, Piepenbrock, Jelle, Urban, Josef
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914193141137408
author Jakubův, Jan
Janota, Mikoláš
Piepenbrock, Jelle
Urban, Josef
author_facet Jakubův, Jan
Janota, Mikoláš
Piepenbrock, Jelle
Urban, Josef
contents In this work we considerably improve the state-of-the-art SMT solving on first-order quantified problems by efficient machine learning guidance of quantifier selection. Quantifiers represent a significant challenge for SMT and are technically a source of undecidability. In our approach, we train an efficient machine learning model that informs the solver which quantifiers should be instantiated and which not. Each quantifier may be instantiated multiple times and the set of the active quantifiers changes as the solving progresses. Therefore, we invoke the ML predictor many times, during the whole run of the solver. To make this efficient, we use fast ML models based on gradient boosting decision trees. We integrate our approach into the state-of-the-art cvc5 SMT solver and show a considerable increase of the system's holdout-set performance after training it on a large set of first-order problems collected from the Mizar Mathematical Library.
format Preprint
id arxiv_https___arxiv_org_abs_2408_14338
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Machine Learning for Quantifier Selection in cvc5
Jakubův, Jan
Janota, Mikoláš
Piepenbrock, Jelle
Urban, Josef
Artificial Intelligence
Machine Learning
Logic in Computer Science
In this work we considerably improve the state-of-the-art SMT solving on first-order quantified problems by efficient machine learning guidance of quantifier selection. Quantifiers represent a significant challenge for SMT and are technically a source of undecidability. In our approach, we train an efficient machine learning model that informs the solver which quantifiers should be instantiated and which not. Each quantifier may be instantiated multiple times and the set of the active quantifiers changes as the solving progresses. Therefore, we invoke the ML predictor many times, during the whole run of the solver. To make this efficient, we use fast ML models based on gradient boosting decision trees. We integrate our approach into the state-of-the-art cvc5 SMT solver and show a considerable increase of the system's holdout-set performance after training it on a large set of first-order problems collected from the Mizar Mathematical Library.
title Machine Learning for Quantifier Selection in cvc5
topic Artificial Intelligence
Machine Learning
Logic in Computer Science
url https://arxiv.org/abs/2408.14338