通过迭代线性规划高效学习平衡符号图

Haruki Yokota, Hiroshi Higashi, Yuichi Tanaka, Gene Cheung
{"title":"通过迭代线性规划高效学习平衡符号图","authors":"Haruki Yokota, Hiroshi Higashi, Yuichi Tanaka, Gene Cheung","doi":"arxiv-2409.07794","DOIUrl":null,"url":null,"abstract":"Signed graphs are equipped with both positive and negative edge weights,\nencoding pairwise correlations as well as anti-correlations in data. A balanced\nsigned graph has no cycles of odd number of negative edges. Laplacian of a\nbalanced signed graph has eigenvectors that map simply to ones in a\nsimilarity-transformed positive graph Laplacian, thus enabling reuse of\nwell-studied spectral filters designed for positive graphs. We propose a fast\nmethod to learn a balanced signed graph Laplacian directly from data.\nSpecifically, for each node $i$, to determine its polarity $\\beta_i \\in\n\\{-1,1\\}$ and edge weights $\\{w_{i,j}\\}_{j=1}^N$, we extend a sparse inverse\ncovariance formulation based on linear programming (LP) called CLIME, by adding\nlinear constraints to enforce ``consistent\" signs of edge weights\n$\\{w_{i,j}\\}_{j=1}^N$ with the polarities of connected nodes -- i.e.,\npositive/negative edges connect nodes of same/opposing polarities. For each LP,\nwe adapt projections on convex set (POCS) to determine a suitable CLIME\nparameter $\\rho > 0$ that guarantees LP feasibility. We solve the resulting LP\nvia an off-the-shelf LP solver in $\\mathcal{O}(N^{2.055})$. Experiments on\nsynthetic and real-world datasets show that our balanced graph learning method\noutperforms competing methods and enables the use of spectral filters and graph\nconvolutional networks (GCNs) designed for positive graphs on signed graphs.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Efficient Learning of Balanced Signed Graphs via Iterative Linear Programming\",\"authors\":\"Haruki Yokota, Hiroshi Higashi, Yuichi Tanaka, Gene Cheung\",\"doi\":\"arxiv-2409.07794\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Signed graphs are equipped with both positive and negative edge weights,\\nencoding pairwise correlations as well as anti-correlations in data. A balanced\\nsigned graph has no cycles of odd number of negative edges. Laplacian of a\\nbalanced signed graph has eigenvectors that map simply to ones in a\\nsimilarity-transformed positive graph Laplacian, thus enabling reuse of\\nwell-studied spectral filters designed for positive graphs. We propose a fast\\nmethod to learn a balanced signed graph Laplacian directly from data.\\nSpecifically, for each node $i$, to determine its polarity $\\\\beta_i \\\\in\\n\\\\{-1,1\\\\}$ and edge weights $\\\\{w_{i,j}\\\\}_{j=1}^N$, we extend a sparse inverse\\ncovariance formulation based on linear programming (LP) called CLIME, by adding\\nlinear constraints to enforce ``consistent\\\" signs of edge weights\\n$\\\\{w_{i,j}\\\\}_{j=1}^N$ with the polarities of connected nodes -- i.e.,\\npositive/negative edges connect nodes of same/opposing polarities. For each LP,\\nwe adapt projections on convex set (POCS) to determine a suitable CLIME\\nparameter $\\\\rho > 0$ that guarantees LP feasibility. We solve the resulting LP\\nvia an off-the-shelf LP solver in $\\\\mathcal{O}(N^{2.055})$. Experiments on\\nsynthetic and real-world datasets show that our balanced graph learning method\\noutperforms competing methods and enables the use of spectral filters and graph\\nconvolutional networks (GCNs) designed for positive graphs on signed graphs.\",\"PeriodicalId\":501034,\"journal\":{\"name\":\"arXiv - EE - Signal Processing\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - EE - Signal Processing\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07794\",\"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 - EE - Signal Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07794","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

有符号图既有正边权重,也有负边权重,可对数据中的成对相关性和反相关性进行编码。平衡有符号图没有奇数负边的循环。平衡有符号图的拉普拉矢具有特征向量,这些特征向量可以简单地映射到相似性变换后的正图拉普拉矢中的特征向量,因此可以重复使用为正图设计的、经过深入研究的光谱滤波器。我们提出了一种直接从数据中学习平衡符号图拉普拉奇的快速方法。具体来说,对于每个节点 $i$,为了确定其极性 $beta_i \in\{-1,1\}$ 和边权重 $\{w_{i,j}\}_{j=1}^N$,我们扩展了一种基于线性规划(LP)的稀疏逆协方差公式,称为 CLIME、通过添加线性约束来强制边缘权重${w_{i,j}\}_{j=1}^N$的符号与相连节点的极性 "一致"--也就是说,边缘权重${w_{i,j}\}_{j=1}^N$与相连节点的极性 "一致"。e.,正/负边缘连接极性相同/相反的节点。对于每个 LP,我们都采用凸集投影法(POCS)来确定一个合适的 CLIME 参数 $\rho > 0$,以保证 LP 的可行性。我们使用现成的 LP 求解器在 $\mathcal{O}(N^{2.055})$ 内求解得到的 LP。在合成数据集和现实世界数据集上的实验表明,我们的平衡图学习方法优于其他竞争方法,并能在有符号图上使用为正图设计的光谱滤波器和图卷积网络(GCN)。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Efficient Learning of Balanced Signed Graphs via Iterative Linear Programming
Signed graphs are equipped with both positive and negative edge weights, encoding pairwise correlations as well as anti-correlations in data. A balanced signed graph has no cycles of odd number of negative edges. Laplacian of a balanced signed graph has eigenvectors that map simply to ones in a similarity-transformed positive graph Laplacian, thus enabling reuse of well-studied spectral filters designed for positive graphs. We propose a fast method to learn a balanced signed graph Laplacian directly from data. Specifically, for each node $i$, to determine its polarity $\beta_i \in \{-1,1\}$ and edge weights $\{w_{i,j}\}_{j=1}^N$, we extend a sparse inverse covariance formulation based on linear programming (LP) called CLIME, by adding linear constraints to enforce ``consistent" signs of edge weights $\{w_{i,j}\}_{j=1}^N$ with the polarities of connected nodes -- i.e., positive/negative edges connect nodes of same/opposing polarities. For each LP, we adapt projections on convex set (POCS) to determine a suitable CLIME parameter $\rho > 0$ that guarantees LP feasibility. We solve the resulting LP via an off-the-shelf LP solver in $\mathcal{O}(N^{2.055})$. Experiments on synthetic and real-world datasets show that our balanced graph learning method outperforms competing methods and enables the use of spectral filters and graph convolutional networks (GCNs) designed for positive graphs on signed graphs.
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
期刊最新文献
Blind Deconvolution on Graphs: Exact and Stable Recovery End-to-End Learning of Transmitter and Receiver Filters in Bandwidth Limited Fiber Optic Communication Systems Atmospheric Turbulence-Immune Free Space Optical Communication System based on Discrete-Time Analog Transmission User Subgrouping in Scalable Cell-Free Massive MIMO Multicasting Systems Covert Communications Without Pre-Sharing of Side Information and Channel Estimation Over Quasi-Static Fading Channels
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1