LinSATNet: The Positive Linear Satisfiability Neural Networks

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Runzhong, Zhang, Yunhao, Guo, Ziao, Chen, Tianyi, Yang, Xiaokang, Yan, Junchi
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913436748742656
author Wang, Runzhong
Zhang, Yunhao
Guo, Ziao
Chen, Tianyi
Yang, Xiaokang
Yan, Junchi
author_facet Wang, Runzhong
Zhang, Yunhao
Guo, Ziao
Chen, Tianyi
Yang, Xiaokang
Yan, Junchi
contents Encoding constraints into neural networks is attractive. This paper studies how to introduce the popular positive linear satisfiability to neural networks. We propose the first differentiable satisfiability layer based on an extension of the classic Sinkhorn algorithm for jointly encoding multiple sets of marginal distributions. We further theoretically characterize the convergence property of the Sinkhorn algorithm for multiple marginals. In contrast to the sequential decision e.g.\ reinforcement learning-based solvers, we showcase our technique in solving constrained (specifically satisfiability) problems by one-shot neural networks, including i) a neural routing solver learned without supervision of optimal solutions; ii) a partial graph matching network handling graphs with unmatchable outliers on both sides; iii) a predictive network for financial portfolios with continuous constraints. To our knowledge, there exists no one-shot neural solver for these scenarios when they are formulated as satisfiability problems. Source code is available at https://github.com/Thinklab-SJTU/LinSATNet
format Preprint
id arxiv_https___arxiv_org_abs_2407_13917
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LinSATNet: The Positive Linear Satisfiability Neural Networks
Wang, Runzhong
Zhang, Yunhao
Guo, Ziao
Chen, Tianyi
Yang, Xiaokang
Yan, Junchi
Artificial Intelligence
Optimization and Control
Encoding constraints into neural networks is attractive. This paper studies how to introduce the popular positive linear satisfiability to neural networks. We propose the first differentiable satisfiability layer based on an extension of the classic Sinkhorn algorithm for jointly encoding multiple sets of marginal distributions. We further theoretically characterize the convergence property of the Sinkhorn algorithm for multiple marginals. In contrast to the sequential decision e.g.\ reinforcement learning-based solvers, we showcase our technique in solving constrained (specifically satisfiability) problems by one-shot neural networks, including i) a neural routing solver learned without supervision of optimal solutions; ii) a partial graph matching network handling graphs with unmatchable outliers on both sides; iii) a predictive network for financial portfolios with continuous constraints. To our knowledge, there exists no one-shot neural solver for these scenarios when they are formulated as satisfiability problems. Source code is available at https://github.com/Thinklab-SJTU/LinSATNet
title LinSATNet: The Positive Linear Satisfiability Neural Networks
topic Artificial Intelligence
Optimization and Control
url https://arxiv.org/abs/2407.13917