Abstract
Multi-target regression (MTR) refers to learning multiple relevant regression tasks simultaneously. Although much progress has been made in multi-target regression, there are still two challenging issues, that is, how to model the underlying relationships between input features and output targets, and how to explore inter-target dependencies. In this study, an effective algorithm named LLIC is proposed; it learns local instance correlations to reveal the relationships between features and output targets, and inter-target dependencies. First, an eminent instance selection method is adapted to directly work with multi-target data, constructing a collection of local instances for each instance. Then, in order to exploit the relationships between input features and output targets, and reveal inter-target dependencies, the collection of local instances is divided into two spaces, that is, a feature space and a target space. Implicit features of input features and targets are obtained in a statistical way. Finally, a final prediction model for each output target is trained on an expanded input space where the implicit features are treated as additional input variables. Extensive experiments on 18 benchmark datasets demonstrate that our proposed LLIC method can achieve competitive performance against representative state-of-the-art multi-target regression methods.
Similar content being viewed by others
Notes
http://mulan.sourceforge.net/datasets-mtr.html
References
Breskvar M, Kocev D, Dzeroski S (2018) Ensembles for multi-target regression with random output selections. Mach Learn 107(11):1673–1709
Spyromitros-Xioufis E, Sechidis K, Vlahavas (2020) Multi-target regression via output space quantization. Machine Learning. arXiv:2003.09896
Petkovic M, Kocev D, Dzeroski S (2020) Feature ranking for multi-target regression. Mach Learn 109:1179–1204
Wang J, Chen Z, Sun K, Li H, Deng X (2019) Multi-target regression via target specific features. Knowl-Based Syst 170:70–78
Wang Y, Wipf DP, Ling Q, Chen W, Wassell IJ (2015) Multi-task learning for subspace segmentation. In: Proceedings of the 32nd international conference on machine learning (ICML), pp 1209–1217
Xiong T, Bao YK, Hu ZY (2014) Multiple-output support vector regression with a firefly algorithm for interval-valued stock price index forecasting. Knowl-Based Syst 55:87–100
Hadavandi E, Shahrabi J, Shamshirband S (2015) A novel Boosted-neural network ensemble for modeling multi-target regression problems. Eng Appl Artif Intel 45:204–219
Stojanova D, Ceci M, Appice A, Dzeroski S (2012) Network regression with predictive clustering trees. Data Min Knowl Disc 25(2):378–413
Yan Y, Ricci E, Subramanian R, Liu GW, Lanz O, Sebe N (2016) A multi-task learning framework for head pose estimation under target motion. IEEE Trans Pattern Anal Mach Intell 38(6):1070–1083
Zhen XT, Wang ZJ, Ali I, Bhaduri M, Chan I, Li S (2016) Multi-scale deep networks and regression forests for direct bi-ventricular volume estimation. Med Image Anal 30:120–129
Wang X, Zhen X, Li Q, Shen D, Huang H (2018) Cognitive assessment prediction in alzheimer’s disease by multi-layer multi-target regression. Neuroinformatics 16:285–294
Zhen XT, Yu MY, He XF, Li S (2018) Multi-target regression via robust low-rank learning. IEEE Trans Pattern Anal Mach Intell 40(2):497–504
Borchani H, Varando G, Bielza C, Larranaga P (2015) A survey on multi-output regression. Wiley Interdiscip Rev Data Min Knowl Discov 5(5):216–233
Lapin M, Hein M, Schiele B (2018) Analysis and optimization of loss functions for multiclass, top-k, and multilabel classification. IEEE Trans Pattern Anal Mach Intell 40(7):1533–1554
Osojnik A, Panov P, Dzeroski S (2017) Multi-label classification via multi-target regression on data streams. Mach Learn 106(6):745–770
Spyromitros Xioufis E, Tsoumakas G, Groves W, Vlahavas IP (2016) Multi-target regression via input space expansion: treating targets as inputs. Mach Learn 104(1):55–98
Zhen XT, Yu MY, Zheng F, Ben Nachum I, Bhaduri M, Laidley DT, Li S (2018) Multitarget sparse latent regression. IEEE Trans Neural Netw Learning Syst 29(5):1575–1586
Melki G, Cano A, Kecman V, Ventura S (2017) Multi-target support vector regression via correlation regressor chains. Inform Sci 415:53–69
Read J, Hollmen J (2014) A deep interpretation of classifier chains. Adv Intell Data Anal, 251–262
Tsoumakas G, Vlahavas IP (2007) Random k-Labelsets: an ensemble method for multilabel classification. European Conference on Machine Learning, 406–417
Tsoumakas G, Spyromitros Xioufis E, Vrekou A, Vlahavas IP (2014) Multi-target regression via random linear target combinations. European Conference Machine Learning and Knowledge Discovery in Databases, 225–240
Zhang Z, Gu J (2020) Facial affect recognition in the wild using multi-task learning convolutional network. Computer Vision and Pattern Recognition. arXiv:2002.00606
Su F, Shang HY, Wang JY (2019) Low-rank deep convolutional neural network for multi-task learning. Comput Intell Neurosci 2019:1–10
Rai P, Kumar A, Daume H (2012) Simultaneously leveraging output and task structures for multiple-output regression. Advances in Neural Information Processing Systems, 3194–3202
Zhang Y, Yeung DY (2012) A convex formulation for learning task relationships in multi-task learning. arXiv:1203.3536
Alvarez MA, Rosasco L, Lawrence ND (2011) Kernels for vector-valued functions: a review. Found Trends Mach Learn 4(3):195–266
Aho T, Zenko B, Dzeroski S, Elomaa T (2012) Multi-target regression with rule ensembles. J Mach Learn Res 13(1):2367–2407
Osojnik A, Panov P, Dzeroski S (2018) Tree-based methods for online multi-target regression. J Intell Inform Syst 50:315–339
Levatic J, Ceci M, Kocev D, Dzeroski S (2017) Self-training for multi-target regression with tree ensembles. Knowl-Based Syst 123:41–60
Stepisnik T, Osojnik A, Dzeroski S, Kocev D (2020) Option predictive clustering trees for multi-target regression. Comput Sci Inform Syst 17:6–6
Mastelini SM, Da Costa VGT, Santana EJ, Nakaro FK, Guido RC, Cerri R, Barbon S Jr (2019) Multi-output tree chaining: an interpretative modelling and lightweight multi-target approach. J Signal Process Syst 91:191–215
Demsar J (2006) Statistical comparisons of classifiers over multiple data sets. J Mach Learn Res 7:1–30
Acknowledgements
This work is supported by the National Natural Science Foundation of China (Grant No.61806033), and Natural Science Foundation of Chongqing (Grant No. cstc2019jcyj-msxmX0021)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Sun, K., Deng, M., Li, H. et al. Learning local instance correlations for multi-target regression. Appl Intell 51, 6124–6135 (2021). https://doi.org/10.1007/s10489-020-02112-5
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-020-02112-5