针对最大独立集问题的无数据四元神经网络

Ismail Alkhouri, Cedric Le Denmat, Yingjie Li, Cunxi Yu, Jia Liu, Rongrong Wang, Alvaro Velasquez
{"title":"针对最大独立集问题的无数据四元神经网络","authors":"Ismail Alkhouri, Cedric Le Denmat, Yingjie Li, Cunxi Yu, Jia Liu, Rongrong Wang, Alvaro Velasquez","doi":"arxiv-2406.19532","DOIUrl":null,"url":null,"abstract":"Combinatorial Optimization (CO) plays a crucial role in addressing various\nsignificant problems, among them the challenging Maximum Independent Set (MIS)\nproblem. In light of recent advancements in deep learning methods, efforts have\nbeen directed towards leveraging data-driven learning approaches, typically\nrooted in supervised learning and reinforcement learning, to tackle the NP-hard\nMIS problem. However, these approaches rely on labeled datasets, exhibit weak\ngeneralization, and often depend on problem-specific heuristics. Recently,\nReLU-based dataless neural networks were introduced to address combinatorial\noptimization problems. This paper introduces a novel dataless quadratic neural\nnetwork formulation, featuring a continuous quadratic relaxation for the MIS\nproblem. Notably, our method eliminates the need for training data by treating\nthe given MIS instance as a trainable entity. More specifically, the graph\nstructure and constraints of the MIS instance are used to define the structure\nand parameters of the neural network such that training it on a fixed input\nprovides a solution to the problem, thereby setting it apart from traditional\nsupervised or reinforcement learning approaches. By employing a gradient-based\noptimization algorithm like ADAM and leveraging an efficient off-the-shelf GPU\nparallel implementation, our straightforward yet effective approach\ndemonstrates competitive or superior performance compared to state-of-the-art\nlearning-based methods. Another significant advantage of our approach is that,\nunlike exact and heuristic solvers, the running time of our method scales only\nwith the number of nodes in the graph, not the number of edges.","PeriodicalId":501216,"journal":{"name":"arXiv - CS - Discrete Mathematics","volume":"38 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Dataless Quadratic Neural Networks for the Maximum Independent Set Problem\",\"authors\":\"Ismail Alkhouri, Cedric Le Denmat, Yingjie Li, Cunxi Yu, Jia Liu, Rongrong Wang, Alvaro Velasquez\",\"doi\":\"arxiv-2406.19532\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Combinatorial Optimization (CO) plays a crucial role in addressing various\\nsignificant problems, among them the challenging Maximum Independent Set (MIS)\\nproblem. In light of recent advancements in deep learning methods, efforts have\\nbeen directed towards leveraging data-driven learning approaches, typically\\nrooted in supervised learning and reinforcement learning, to tackle the NP-hard\\nMIS problem. However, these approaches rely on labeled datasets, exhibit weak\\ngeneralization, and often depend on problem-specific heuristics. Recently,\\nReLU-based dataless neural networks were introduced to address combinatorial\\noptimization problems. This paper introduces a novel dataless quadratic neural\\nnetwork formulation, featuring a continuous quadratic relaxation for the MIS\\nproblem. Notably, our method eliminates the need for training data by treating\\nthe given MIS instance as a trainable entity. More specifically, the graph\\nstructure and constraints of the MIS instance are used to define the structure\\nand parameters of the neural network such that training it on a fixed input\\nprovides a solution to the problem, thereby setting it apart from traditional\\nsupervised or reinforcement learning approaches. By employing a gradient-based\\noptimization algorithm like ADAM and leveraging an efficient off-the-shelf GPU\\nparallel implementation, our straightforward yet effective approach\\ndemonstrates competitive or superior performance compared to state-of-the-art\\nlearning-based methods. Another significant advantage of our approach is that,\\nunlike exact and heuristic solvers, the running time of our method scales only\\nwith the number of nodes in the graph, not the number of edges.\",\"PeriodicalId\":501216,\"journal\":{\"name\":\"arXiv - CS - Discrete Mathematics\",\"volume\":\"38 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-06-27\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Discrete Mathematics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2406.19532\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Discrete Mathematics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2406.19532","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

组合优化(Combinatorial Optimization,CO)在解决各种重大问题中发挥着至关重要的作用,其中包括极具挑战性的最大独立集(MIS)问题。鉴于深度学习方法的最新进展,人们开始努力利用数据驱动的学习方法(通常以监督学习和强化学习为基础)来解决 NP-hardMIS问题。然而,这些方法依赖于有标签的数据集,表现出弱泛化性,而且往往依赖于特定问题的启发式方法。最近,基于 ReLU 的无数据神经网络被引入解决组合优化问题。本文介绍了一种新颖的无数据二次神经网络公式,其特点是对 MIS 问题进行连续的二次松弛。值得注意的是,我们的方法将给定的 MIS 实例视为可训练实体,从而消除了对训练数据的需求。更具体地说,MIS 实例的图结构和约束条件被用来定义神经网络的结构和参数,这样在固定输入上对其进行训练就能得到问题的解决方案,从而使其有别于传统的监督或强化学习方法。通过采用基于梯度的优化算法(如 ADAM),并利用高效的现成 GPU 并行执行,我们的方法简单而有效,与基于学习的先进方法相比,性能具有竞争力或更优越。我们方法的另一个显著优势是,与精确求解器和启发式求解器不同,我们方法的运行时间只与图中的节点数量而不是边的数量有关。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Dataless Quadratic Neural Networks for the Maximum Independent Set Problem
Combinatorial Optimization (CO) plays a crucial role in addressing various significant problems, among them the challenging Maximum Independent Set (MIS) problem. In light of recent advancements in deep learning methods, efforts have been directed towards leveraging data-driven learning approaches, typically rooted in supervised learning and reinforcement learning, to tackle the NP-hard MIS problem. However, these approaches rely on labeled datasets, exhibit weak generalization, and often depend on problem-specific heuristics. Recently, ReLU-based dataless neural networks were introduced to address combinatorial optimization problems. This paper introduces a novel dataless quadratic neural network formulation, featuring a continuous quadratic relaxation for the MIS problem. Notably, our method eliminates the need for training data by treating the given MIS instance as a trainable entity. More specifically, the graph structure and constraints of the MIS instance are used to define the structure and parameters of the neural network such that training it on a fixed input provides a solution to the problem, thereby setting it apart from traditional supervised or reinforcement learning approaches. By employing a gradient-based optimization algorithm like ADAM and leveraging an efficient off-the-shelf GPU parallel implementation, our straightforward yet effective approach demonstrates competitive or superior performance compared to state-of-the-art learning-based methods. Another significant advantage of our approach is that, unlike exact and heuristic solvers, the running time of our method scales only with the number of nodes in the graph, not the number of edges.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信