Multi-Target regression via low rank matrix factorization for data with label noise

LIU Zhigang; LIU Senze

doi:10.19304/J.ISSN1000-7180.2021.0787

您当前的位置：

首页 >

文章列表页 >

Multi-Target regression via low rank matrix factorization for data with label noise

更新时间：2026-09-04

- Multi-Target regression via low rank matrix factorization for data with label noise
- Vol. 39, Issue 3, Pages: 50-57(2022)
- 作者机构：
  
  1. 无锡职业技术学院机构控制技术学院,江苏,无锡,214121
  2. 南京大学计算机科学与技术系,江苏,南京,210023
- 作者简介：
- 基金信息：
- DOI：10.19304/J.ISSN1000-7180.2021.0787
  CLC：
- Published：2022
- 稿件说明：
移动端阅览
刘志刚, 刘森泽. 存在标签噪音的数据中基于低秩矩阵分解的多输出回归[J]. 微电子学与计算机, 2022, 39(3): 50-57.

LIU Zhigang, LIU Senze. Multi-Target regression via low rank matrix factorization for data with label noise[J]. 2022, 39(3): 50-57.
刘志刚, 刘森泽. 存在标签噪音的数据中基于低秩矩阵分解的多输出回归[J]. 微电子学与计算机, 2022, 39(3): 50-57. DOI： 10.19304/J.ISSN1000-7180.2021.0787.

LIU Zhigang, LIU Senze. Multi-Target regression via low rank matrix factorization for data with label noise[J]. 2022, 39(3): 50-57. DOI： 10.19304/J.ISSN1000-7180.2021.0787.

摘要

多输出回归是指针对一组输入变量来估计其对应的多个连续属性值，其在数据挖掘领域有着广泛的应用.当前关于多输出回归任务的研究都是基于标签值准确的假设下实现的.然而在实际情况中，数据集的部分标签可能并不准确，即部分标签存在一定的噪声.在这种情况下，传统多输出回归方法性能较差.为了解决上述情况下的多输出回归问题，利用大数据中数据样本大的特点来提炼各个标签间的相关性，从而利用标签间的相关矩阵重构标签.由于多输出问题中的标签个数通常较多，因此可以一定程度上稀释掉部分标签的噪声干扰.此外，利用低秩矩阵分解对上述思路建立数学优化问题，并在此基础上引入核技巧以提升模型的非线性拟合能力.最后，采用非凸近似的手段求解该优化问题，从而保证了多输出回归模型的预测性能.实验18个数据集上同现有的6种多输出回归方法进行了比较，提出的方法在样本量较大的场景下性能优势较为明显.

Abstract

Multiple-target regression refers to a set of input variables to estimate its corresponding multiple continuous attribute values

which has a wide range of applications in the field of data mining. The current research on multi-target regression tasks is based on the assumption that the label value is accurate. However

in actual situations

some labels of the data set may not be accurate

that is

some labels have certain noise. In this case

the traditional multi-target regression methods usually cannot achieve good results. In order to solve the multi-target regression problem in the above situation

the characteristics of large data samples in big data are used to refine the correlation between labels

and then the correlation matrix between labels is used to reconstruct the labels. Since the number of labels in the multi-target problem is usually large

the noise interference of some labels can be diluted to a certain extent. In addition

low-rank matrix decomposition is used to establish a mathematical optimization problem for the above ideas

and on this basis

kernel techniques are introduced to improve the nonlinear fitting ability of the model. Finally

a non-convex approximation method is used to solve the optimization problem

thereby ensuring the prediction performance of the multi-target regression model. The experiment is compared with 6 existing multi-target regression models on 18 datasets. The method proposed in this paper has obvious performance advantages in scenarios with large sample sizes.

关键词

Keywords

references

TSOUMAKAS G, SPYROMITROS-XIOUFIS E, VILCEK J, et al. MULAN: A Java library for multi-label learning[J]. Journal of Machine Learning Research, 2011(12): 2411-2414. DOI: 2411

SÁNCHEZ-FERNáNDEZ M, DE-PRADO-CUMPLIDO M, ARE-NAS-GARCíA J, et al. SVM multiregression for nonlinear channel estimation in multiple-input multiple-output systems[J]. IEEE Transactions on Signal Processing, 2004, 52(8): 2298-2307. DOI: 10.1109/TSP.2004.831028.

RAI P, KUMAR A, DAUME H. Simultaneously leveraging output and task structures for mul-tiple-output regression[C]//Proceedings of the 25th International Conference on Neural Information Processing Systems. New York: Curran Associates Inc., 2012: 3185-3193.

RANJAN R, PATEL V M, CHELLAPPA R. HyperFace: a deep multi-task learning framework for face detection, landmark localization, pose es-timation, and gender recognition[J]. IEEE Trans-actions on Pattern Analysis and Machine Intelli-gence, 2017, 41(1): 121-135. DOI: 10.1109/TPAMI.2017.2781233.

WANG X, WANG F, HU J Y. A multi-task learning framework for joint disease risk prediction and comorbidity discovery[C]//2014 22nd Internation-al Conference on Pattern Recognition. Stockholm: IEEE, 2014: 220-225. DOI: 10.1109/ICPR.2014.47.

RODRÍGUEZ J J, JUEZ-GIL M, LÓPEZ-NOZAL C, et al. Rotation Forest for multi-target regression[J]. International Journal of Machine Learning and Cybernetics, 202 1: 1-26. DOI: 10. 1007/s 13042-02 1-0 1329- 1.

MELKI G, CANO A, KECMAN V, et al. Mul-ti-target support vector regression via correlation regressorchains[J]. Information Sciences, 2017, 415-416: 53-69. DOI: 10.1016/j.ins.2017.06.017.

EVGENIOU T, MICCHELLI C A, PONTIL M. Learning multiple tasks with kernel methods[J]. Journal of Machine Learning Research, 2005(6): 615-637. DOI: 615

BREIMAN L, FRIEDMAN J H. Predicting multi-variate responses in multiple linear regression[J]. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 1997, 59(1): 3-54. DOI: 10.1111/1467-9868.00054.

SPYROMITROS-XIOUFIS E, TSOUMAKAS G, GROVES W, et al. Multi-target regression via in-put space expansion: treating targets as inputs[J]. Machine Learning, 2016, 104(1): 55-98. DOI: 10.1007/s10994-016- 5546-z.

ÁLVAREZM A, ROSASCO L, LAWRENCE N D. Kernels for vector-valued functions: areview[J]. Foundations and Trendsá in Machine Learning, 2012, 4(3): 195-266. DOI: 195

HARA K, CHELLAPPA R. Growing regression forests by classification: applications to object pose estimation[C]//Proceedings of the 13th European Conference on Computer Vi-sion. Zurich: Springer, 2014: 552-567. DOI: 10.1007/978-3-319-10605-2_36.

ZHANG M L, ZHOU Z H. A review on multi-label learning algorithms[J]. IEEETransactions on Knowledge and Data Engineering, 2014, 26(8): 1819-1837. DOI: 10.1109/TKDE.2013.39.

Cambridge SHAWE-TAYLOR J, CRISTIANINI N. Kernel methods for pattern analysis[M]. Cambridge: Cambridge University Press, 2004.

TSOUMAKAS G, SPYROMITROS-XIOUFIS E, VREKOU A, et al. Multi-target regression via random linear target combinations[C]//Joint Eu-ropean Conference on Machine Learning and Knowledge Discovery in Databases. Nancy: Springer, 2014: 225-240. DOI: 10.1007/978-3-662-44845-8_15.

XIA R, LIU Y. A multi-task learning framework for emotion recognition using 2D continuous space[J]. IEEE Transactions on Affective Computing, 2017, 8(1): 3-14. DOI: 10.1109/TAFFC.2015.2512598.

EVGENIOU A, PONTIL M. Multi-task feature learning[J]. Advances in Neural Information Processing Systems, 2007(19): 41. DOI: 41

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A feature-oriented context understanding and slice analysis network for tuberculosis detection

A review of large model development: technological evolution from BERT to Deepseek

Implementation and verification of a dual-chain power gating strategy for low power design

Microservices anomaly detection based on the correlation of monitoring metrics

A high-efficiency charge pump for ultra-low voltage boost systems

Related Author

No data

Related Institution

No data

AI问答

Postal code：100079
Tel：（010）53879206 Email：tmw@bjxintong.com.cn
Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09082226号-64 京公网安备11010602201714号
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰