{"title":"无向加权网络的交替非负最小二乘法--纳入正则化的对称潜因子分析","authors":"","doi":"10.1016/j.neucom.2024.128440","DOIUrl":null,"url":null,"abstract":"<div><p>An <u>U</u>ndirected <u>W</u>eighted <u>N</u>etwork (UWN) can be precisely quantified as an adjacency matrix whose inherent characteristics are fully considered in a <u>S</u>ymmetric <u>N</u>onnegative <u>L</u>atent <u>F</u>actor (SNLF) model for its good representation accuracy. However, an SNLF model uses a sole latent factor matrix to precisely describe the topological characteristic of a UWN, i.e., symmetry, thereby impairing its representation learning ability. Aiming at addressing this issue, this paper proposes an <u>A</u>lternating nonnegative least squares-incorporated Regularized <u>S</u>ymmetric <u>L</u>atent factor analysis (ARSL) model. First of all, equation constraints composed of multiple matrices are built in its learning objective for well describing the symmetry of a UWN. Note that it adopts an <em>L</em><sub><em>2</em></sub>-norm-based regularization scheme to relax such constraints for making such a symmetry-aware learning objective solvable. Then, it designs an alternating nonnegative least squares-incorporated algorithm for optimizing its parameters efficiently. Empirical studies on four UWNs demonstrate that an ARSL model outperforms the state-of-the-art models in terms of representation accuracy, as well as achieves promising computational efficiency.</p></div>","PeriodicalId":19268,"journal":{"name":"Neurocomputing","volume":null,"pages":null},"PeriodicalIF":5.5000,"publicationDate":"2024-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Alternating nonnegative least squares-incorporated regularized symmetric latent factor analysis for undirected weighted networks\",\"authors\":\"\",\"doi\":\"10.1016/j.neucom.2024.128440\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>An <u>U</u>ndirected <u>W</u>eighted <u>N</u>etwork (UWN) can be precisely quantified as an adjacency matrix whose inherent characteristics are fully considered in a <u>S</u>ymmetric <u>N</u>onnegative <u>L</u>atent <u>F</u>actor (SNLF) model for its good representation accuracy. However, an SNLF model uses a sole latent factor matrix to precisely describe the topological characteristic of a UWN, i.e., symmetry, thereby impairing its representation learning ability. Aiming at addressing this issue, this paper proposes an <u>A</u>lternating nonnegative least squares-incorporated Regularized <u>S</u>ymmetric <u>L</u>atent factor analysis (ARSL) model. First of all, equation constraints composed of multiple matrices are built in its learning objective for well describing the symmetry of a UWN. Note that it adopts an <em>L</em><sub><em>2</em></sub>-norm-based regularization scheme to relax such constraints for making such a symmetry-aware learning objective solvable. Then, it designs an alternating nonnegative least squares-incorporated algorithm for optimizing its parameters efficiently. Empirical studies on four UWNs demonstrate that an ARSL model outperforms the state-of-the-art models in terms of representation accuracy, as well as achieves promising computational efficiency.</p></div>\",\"PeriodicalId\":19268,\"journal\":{\"name\":\"Neurocomputing\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":5.5000,\"publicationDate\":\"2024-08-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Neurocomputing\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0925231224012116\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neurocomputing","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0925231224012116","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
Alternating nonnegative least squares-incorporated regularized symmetric latent factor analysis for undirected weighted networks
An Undirected Weighted Network (UWN) can be precisely quantified as an adjacency matrix whose inherent characteristics are fully considered in a Symmetric Nonnegative Latent Factor (SNLF) model for its good representation accuracy. However, an SNLF model uses a sole latent factor matrix to precisely describe the topological characteristic of a UWN, i.e., symmetry, thereby impairing its representation learning ability. Aiming at addressing this issue, this paper proposes an Alternating nonnegative least squares-incorporated Regularized Symmetric Latent factor analysis (ARSL) model. First of all, equation constraints composed of multiple matrices are built in its learning objective for well describing the symmetry of a UWN. Note that it adopts an L2-norm-based regularization scheme to relax such constraints for making such a symmetry-aware learning objective solvable. Then, it designs an alternating nonnegative least squares-incorporated algorithm for optimizing its parameters efficiently. Empirical studies on four UWNs demonstrate that an ARSL model outperforms the state-of-the-art models in terms of representation accuracy, as well as achieves promising computational efficiency.
期刊介绍:
Neurocomputing publishes articles describing recent fundamental contributions in the field of neurocomputing. Neurocomputing theory, practice and applications are the essential topics being covered.