融合标签局部相关性的标签分布学习

doi:10.6040/j.issn.1671-9352.4.2021.196

摘要/Abstract

摘要： 提出了一种融合标签局部相关性的标签分布学习(label distribution learning, LDL)算法,该算法分为3个阶段。初始预测阶段构建多层神经网络模型,将样本的原始特征作为输入、初始预测的标签分布作为输出;局部矫正阶段首先利用k-means聚类算法获得不同类所描述的局部信息,然后针对不同类计算对应的协方差矩阵,利用该矩阵来矫正初始预测的标签分布,获得每个类对应的矫正标签分布;标签融合阶段对矫正后的标签分布进行加权,再与初始预测的标签分布进行融合,得到最终的预测分布。在8个公开数据集上与9种常用的LDL算法进行对比实验,结果表明本文的算法能较好地描述标签局部相关性,在多个主流评估指标上排名靠前。

关键词: 标签分布学习, 标签相关性, 标签融合, k-means

Abstract: This paper proposes an LDL algorithm, which integrates the local correlation of labels. The algorithm is divided into three stages. In the initial prediction stage, this paper constructs a multi-layer neural network, which takes the original features as the input and the initial prediction label distribution as the outputs. In the local correction stage, first we use k-means to obtain the local information described by different clusters. Then, for each class, we calculate the corresponding covariance matrix, and finally use this matrix to correct the initial predicted label distribution to obtain the corrected label distribution. In the label fusion stage, the corrected label distribution is weighted, and then fused with the initial predicted label distribution to obtain the final predicted distribution. Compared with night popular LDL algorithms, experiments were conducted on eight different public datasets. The results show that our algorithm can better describe the local correlation of labels, and ranks higher on mainstream evaluation measures.

Key words: label distribution learning, label correlation, label fusion, k-means

中图分类号:

TP391

容斌元,徐媛媛,吕亚兰,张恒汝. 融合标签局部相关性的标签分布学习[J]. 《山东大学学报(理学版)》, 2022, 57(7): 53-64.

RONG Bin-yuan, XU Yuan-yuan, LÜ Ya-lan, ZHANG Heng-ru. Label distribution learning by fusion of local correlation of labels[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2022, 57(7): 53-64.

参考文献

[1] COURT, SAPP B, TASKAR B. Learning from partial labels[J]. Journal of Machine Learning Research, 2011, 12(42):1501-1536.
[2] ZHANG F W, JIA X Y, LI W W. Tensor-based multi-view label enhancement for multi-label learning[C] //Twenty-Ninth International Joint Conference on Artificial Intelligence and Seventeenth Pacific Rim International Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann, 2020, 3: 2369-2375.
[3] LV Jiaqi, WU Tianran, PENG Chenglun, et al. Compact learning for multi-label classification[J]. Pattern Recognition, 2021, 113:107833.
[4] XU Ning, LIU Yunpeng, GENG Xin, et al. Label enhancement for label distribution learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2021, 33(4):1632-1643.
[5] TAN Chao, CHEN Sheng, GENG Xin, et al. A novel probabilistic label enhancement algorithm for multi-label distribution learning[J/OL]. IEEE Transactions on Knowledge and Data Engineering, 2021. doi: 10.1109/TKDE.2021.3054465.
[6] LING Miaogen, GENG Xin. Indoor crowed counting by mixture of Gaussians label distribution learning[J]. IEEE Transactions on Image Processing, 2019, 28(11):5691-5701.
[7] 黄雨婷, 徐媛媛, 张恒汝, 等. 融合标签结构依赖性的标签分布学习[J]. 南京大学学报(自然科学), 2020, 56(4):524-532. HUANG Yuting, XU Yuanyuan, ZHANG Hengru, et al. Label distribution learning by exploiting structural label dependency[J]. Journal of Nanjing University(Natural Science), 2020, 56(4):524-532.
[8] GENG X, SMITH-MILES K, ZHOU Z H. Facial age estimation by learning from label distributions[C] //Proceedings of the AAAI Conference on Artificial Intelligence. Menlo Park: AAAI, 2010: 451-456.
[9] GENG Xin. Label distribution learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2016, 28(7):1734-1748.
[10] GAO Yongbiao, ZHANG Yu, GENG Xin. Label enhancement for label distribution learning via prior knowledge[C] //Proceedings of the International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann, 2020: 3223-3229.
[11] GAO Binbin, XING Chao, XIE Chenwei, et al. Deep label distribution learning with label ambiguity[J]. IEEE Transactions on Image Processing, 2017, 26(6):2825-2823.
[12] ZHOU Ying, XUE Hui, GENG Xin. Emotion distribution recognition from facial expressions[C] //Proceedings of the ACM International Conference on Multimedia. New York: ACM, 2015:1247-1250.
[13] ZHOU Deyu, ZHANG Xuan, GENG Xin, et al. Emotion distribution learning from texts[C] //Proceedings of the Conference on Empirical Methods in Natural Language Processing. Austin: Association for Computational Linguistics. Stroudsburg: ACL, 2016: 638-647.
[14] REN Tingting, JIA Xiuyi, LI Weiwei, et al. Label distribution learning with label correlations via low-rank approximation[C] //Proceedings of the International Joint Conference on Artificial Intelligence. San Francisco: Morgan Kaufmann, 2019: 3325-3331.
[15] JIA Xiuyi, LI Weiwei, LIU Junyu, et al. Label distribution learning by exploiting label correlations[C] //Proceedings of the AAAI Conference on Artificial Intelligence. Menlo Park: AAAI, 2018: 3310-3317.
[16] JIA Xiuyi, LI Zechao, ZHENG Xiang, et al. Label distribution learning with label correlations on local samples[J]. IEEE Transactions on Knowledge and Data Engineering, 2021, 33(4):1619-1631.
[17] ALOISE D, DESHPANDE A, HANSEN P, et al. NP-hardness of Euclidean sum-of-squares clustering[J]. Machine Learning, 2009, 75(2):24-248.
[18] ELMIRAF, SEYED A S S, YASER D, et al. The dice(sorensen)similarity measures for optimal selection with q-Rung orthopair fuzzy information[C] //International Conference on Intelligent and Fuzzy Systems. Cham: Springer, 2020: 484-493.
[19] TIM E, PETER H. Rényi divergence and Kullback-Leibler divergence[J]. IEEE Transactions on Information Theory, 2014, 60(7):3797-3820.
[20] GENG Xin, XIA Yu. Head pose estimation based on multivariate label distribution[C] //Computer Vision and Pattern Recognition. Piscataway: IEEE, 2014: 1837-1842.
[21] ZHAI Yansheng, DAI Jianhua, SHI Hong. Label distribution learning based on ensemble neural networks[C] //Proceedings of the International Conference on Neural Information Processing. Cham: Springer, 2018: 593-602.
[22] ZHAN GHuiying, ZHANG Yu, GENG Xin. Practical age estimation using deep label distribution learning[J]. Frontiers of Computer Science, 2021, 15(3):153318.
[23] YIN Xinyou, GOUDRIAAN Jan, VOS Jan, et al. A flexible sigmoid function of determinate growth[J]. Annals of botany, 2003, 91(3):361-371.
[24] WANG Feng, LIU Weiyang, LIU Haijun, et al. Additive margin softmax for face verification[J]. IEEE Signal Processing Letters, 2018, 25(7):926-930.
[25] YANG Bo, FU Xiao, HONG Mingyi, et al. Towards k-means-friendly spaces: simultaneous deep learning and clustering[C] //Proceedings of the International Conference on Machine Learning. New York: ACM, 2017: 3861-3870.
[26] ZHANG Geng, ZHANG Chengchang, ZHANG Huayu. Improved k-means algorithm based on density Canopy[J]. Knowledge-based Systems, 2018, 145:289-297.
[27] XIAShuyin, PENG Daowan, MENG Deyu, et al. Ball k-means: fast adaptive clustering with no bounds[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 44(1):87-99.
[28] KANUNGO T, MOUNT D M, NETANYAHU N S, et al. An efficient k-means clustering algorithm: analysis andimplementation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7):881-892.
[29] TUZEL L, PORIKLI F, MEER P. Regioncovariance: a fast descriptor for detection and classification[J]. Lecture Notes in Computer Science, 2006, 3952:589-600.
[30] KINGMA D P, BA J L. Adam: a method for stochasticoptimization[J/OL]. Computer Science, 2014. http://de.arxiv.org/pdf/1412.6980.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed