Abstract
Deep ensemble learning models that combine multiple independent deep learning models with multi-layer processing architectures have proven to be effective techniques for improving the accuracy and robustness of deep learning models. In this paper, we propose a diversified kernel ensemble regression method, which is developed from the well-known kernel ridge regression methods. Motivated by multi-view data modeling ideas, we treat each individual kernel as one view of original data in kernel representation space. Therefore, we develop a deep kernel ensemble ridge regression method in Neural Tangent Kernel (NTK) to address the problem faced by traditional kernel ridge regression methods in finding appropriate types of kernels and their parameters. Specifically, as multiple deep kernel regressors share common information from the multi-view kernel representations, our proposed method is built through a hierarchical modeling method, where deep kernel regressors share a common parameter and also have model-specific parameters in individual regressors, which is further helpful in improving our model’s performance. Furthermore, to achieve better diversified deep kernel representations in our proposed method, the Hilbert-Schmidt Independence Criterion (HSIC) is used to regularize our proposed model. In this way, we can find more diversified kernel representations among multiple kernel ensemble regressors to achieve better generalization performance. Experiments on several classification and regression datasets, such as MNIST, TinyImageNet-200, ORL, Credit and YearPredictionMSD demonstrate that our proposed method can achieve best regression performance than other state-of-the-art methods.
Similar content being viewed by others
Data Availability
The UCI datasets analysed during this study are available in the UCI repository, http://archive.ics.uci.edu/. And other datasets analysed during this study are available from the corresponding author on reasonable request.
References
Borandağ E, Özçift A, Kaygusuz Y (2021) Development of majority vote ensemble feature selection algorithm augmentedwith rank allocation to enhance turkish text categorization. Turk J Electr Eng Comput Sci 29(2):514–530
Yu J, Cai Z, He P, **e G, Ling Q (2022) Multi-model ensemble learning method for human expression recognition. ar**v preprint ar**v:2203.14466
Ahn E, Kumar A, Feng D, Fulham M, Kim J (2019) Unsupervised feature learning with k-means and an ensemble of deep convolutional neural networks for medical image classification. ar**v preprint ar**v:1906.03359
Kazemi S, Minaei Bidgoli B, Shamshirband S, Karimi SM, Ghorbani MA, Chau K-W, Kazem Pour R (2018) Novel genetic-based negative correlation learning for estimating soil temperature. Eng Appl Comput Fluid Mech 12(1):506–516
Wu Y, Liu L, **e Z, Chow K-H, Wei W (2021) Boosting ensemble accuracy by revisiting ensemble diversity metrics. In: Proc IEEE Int Conf Comput Vis Pattern Recognit pp 16469–16477
Bartlett P, Freund Y, Lee WS, Schapire RE (1998) Boosting the margin: A new explanation for the effectiveness of voting methods. Ann Stat 26(5):1651–1686
Breiman L (1996) Bagging predictors. Mach Learn 24(2):123–140
Breiman L (2001) Random forests. Mach Learn 45(1):5–32
Freund Y, Schapire RE (1996) Experiments with a new boosting algorithm. In: Icml, vol 96, pp 148–156. Citeseer
Friedman JH (2001) Greedy function approximation: a gradient boosting machine. Annals of statistics, 1189–1232
Wang R, Kwong S, Wang X, Jia Y (2021) Active k-labelsets ensemble for multi-label classification. Pattern Recogn 109:107583
Wang B, Xue B, Zhang M (2020) Particle swarm optimisation for evolving deep neural networks for image classification by evolving and stacking transferable blocks. In: 2020 IEEE Congr Evol Comput (CEC) pp 1–8. IEEE
Liu B, Gu L, Lu F (2019) Unsupervised ensemble strategy for retinal vessel segmentation. In: Int Conf Med Image Comput Comput Assist Interv, pp 111–119. Springer
Ali F, El-Sappagh S, Islam SR, Kwak D, Ali A, Imran M, Kwak K-S (2020) A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion. Inf Fusion 63:208–222
Zhang W, Jiang J, Shao Y, Cui B (2020) Snapshot boosting: a fast ensemble framework for deep neural networks. Sci China Inf Sci 63(1):1–12
Zhang S, Liu M, Yan J (2020) The diversified ensemble neural network. Adv Neural Inf Process Syst 33:16001–16011
Lakshminarayanan B, Pritzel A, Blundell C (2017) Simple and scalable predictive uncertainty estimation using deep ensembles. Adv Neural Inf Process Syst 30
Bhadra S, Kaski S, Rousu J (2017) Multi-view kernel completion. Mach Learn 106(5):713–739
Khan GA, Hu J, Li T, Diallo B, Zhao Y (2022) Multi-view low rank sparse representation method for three-way clustering. Int J Mach Learn Cybern 13(1):233–253
Jacot A, Gabriel F, Hongler C (2018) Neural tangent kernel: Convergence and generalization in neural networks. Adv Neural Inf Process Syst 31
Gretton A, Bousquet O, Smola A, Schölkopf B (2005) Measuring statistical dependence with hilbert-schmidt norms. In: Int Conf Algorithmic Learning Theory, pp 63–77. Springer
Mukkamala S, Sung AH, Abraham A (2003) Intrusion detection using ensemble of soft computing paradigms. In: Intell Syst Design Appl, pp 239–248. Springer, ???
Chen T, Guestrin C (2016) Xgboost: A scalable tree boosting system. In: Proceed 22nd Acm Sigkdd Int Conf Knowl Discov Data Min, pp 785–794
Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu T-Y (2017) Lightgbm: A highly efficient gradient boosting decision tree. Adv Neural Inf Process Syst 30
Pal M (2005) Random forest classifier for remote sensing classification. Int J Remote Sens 26(1):217–222
Ogunleye A, Wang Q-G (2019) Xgboost model for chronic kidney disease diagnosis. IEEE/ACM Trans Comput Biol Bioinf 17(6):2131–2140
Shi Z, Zhang L, Liu Y, Cao X, Ye Y, Cheng M-M, Zheng G (2018) Crowd counting with deep negative correlation learning. In: Proceed IEEE Conf Comput Vis Pattern Recognit, pp 5382–5390
Xue J, Wang Z, Kong D, Wang Y, Liu X, Fan W, Yuan S, Niu S, Li D (2021) Deep ensemble neural-like p systems for segmentation of central serous chorioretinopathy lesion. Inf Fusion 65:84–94
Lin Z, Chen M, Ma Y (2010) The augmented lagrange multiplier method for exact recovery of corrupted low-rank matrices. ar**v preprint ar**v:1009.5055
Schölkopf B, Smola AJ, Bach F (2002) Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT press, ???
Vorontsov MA, Sivokon VP (1998) Stochastic parallel-gradient-descent technique for high-resolution wave-front phase-distortion correction. JOSA A 15(10):2745–2758
Ma C, Qiu X, Beutel D, Lane N (2023) Gradient-less federated gradient boosting tree with learnable learning rates. In: Proceed 3rd Workshop Mach Learn Syst, pp 56–63
Lalev A, Alexandrova A (2023) Recurrent neural networks for forecasting social processes. In: 2023 Int Conf Big Data Knowl Control Syst Eng (BdKCSE), pp 1–5. IEEE
Wan A, Dunlap L, Ho D, Yin J, Lee S, ** H, Petryk S, Bargal SA, Gonzalez JE (2020) Nbdt: neural-backed decision trees. ar**v preprint ar**v:2004.00221
Luo ZT, Sang H, Mallick B (2022) Bamdt: Bayesian additive semi-multivariate decision trees for nonparametric regression. In: Int Conf Mach Learn, pp 14509–14526. PMLR
Zhou Z-H, Feng J (2019) Deep forest. National Science Review 6(1):74–86
Fang C, Cheng L, Mao Y, Zhang D, Fang Y, Li G, Qi H, Jiao L (2023) Separating noisy samples from tail classes for long-tailed image classification with label noise. IEEE Trans Neural Netw Learn Syst
Fonti V, Belitser E (2017) Feature selection using lasso. VU Amsterdam research paper in business analytics 30:1–25
Fang C, Wang Q, Cheng L, Gao Z, Pan C, Cao Z, Zheng Z, Zhang D (2023) Reliable mutual distillation for medical image segmentation under imperfect annotations. IEEE Trans Med Imaging
Yao J, Han L, Guo G, Zheng Z, Cong R, Huang X, Ding J, Yang K, Zhang D, Han J (2024) Position-based anchor optimization for point supervised dense nuclei detection. Neural Netw 171:159–170
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts of interest
This work was funded in part by the National Natural Science Foundation of China (62376108).
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Liu, Z., Xu, Z., Ebhohimhen Abhadiomhen, S. et al. Diversified deep hierarchical kernel ensemble regression. Multimed Tools Appl (2024). https://doi.org/10.1007/s11042-024-19637-3
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s11042-024-19637-3