Abstract
In deep neural networks, the generalization is a vital evaluation metric. As it contributes to avoid over-fitting, Dropout plays an important role in improving the generalization of deep neural networks. Without fully utilizing the training data and the real-time performance of the networks, traditional Dropout and its variants lack of specificity in the selection of inactivated neurons and the planning of dropout rates, resulting in a weaker performance in enhancing the generalization. Therefore, this paper offers an improved Dropout method. As both the training data and the real-time performance of networks can be quantified by the loss, the method uses the loss of the network prediction to guide the selection of inactivated neurons and the determination of dropout rates. The selection is performed by the genetic algorithm, while the results of the selection are used to plan the dropout rate. In essence, this approach encourages the subset of neurons with the higher loss to be trained so as to increase the robustness of neurons and thus improves the generalization of networks. The experimental results demonstrate that the proposed method achieves better generalization on MiniImageNet and Caltech-256 datasets. Compared with the backbone network, the accuracy improves from 66.56% to 72.95%.
Similar content being viewed by others
Data Availability
Data and materials are available.
Code Availability
Code is available.
References
Achille A, Soatto S (2018) Information dropout: learning optimal representations through noisy computation. IEEE Trans Pattern Anal Mach Intell 40 (12):2897–2905
Ambati LS, El-Gayar O (2021) Human activity recognition: a comparison of machine learning approaches. J Midwest Assoc Inf Syst (JMWAIS) 2021 (1):49
Ba J, Frey B (2013) Adaptive dropout for training deep neural networks. Advances in neural information processing systems 26
Baldi P, Sadowski PJ (2013) Understanding dropout. Advances in neural information processing systems 26
Chattopadhay A, Sarkar A, Howlader P, Balasubramanian VN (2018) Grad-cam++: generalized gradient-based visual explanations for deep convolutional networks. In: 2018 IEEE winter conference on applications of computer vision (WACV), IEEE, pp 839–847
Chen Y, Yi Z (2021) Adaptive sparse dropout: learning the certainty and uncertainty in deep neural networks. Neurocomputing 450:354–361
El-Gayar OF, Ambati LS, Nawar N (2020) Wearables, artificial intelligence, and the future of healthcare, 104–129
Fan X, Zhang S, Tanwisuth K, Qian X, Zhou M (2021) Contextual dropout: an efficient sample-dependent dropout module. arXiv:2103.04181
Feng X, Gao X, Luo L (2021) X-sdd: a new benchmark for hot rolled steel strip surface defects detection. Symmetry 13(4):706
Gal Y, Ghahramani Z (2016) Dropout as a bayesian approximation: representing model uncertainty in deep learning. In: International conference on machine learning, PMLR, pp 1050–1059
Gao W, Zhou Z-H (2016) Dropout rademacher complexity of deep neural networks. Sci China Inf Sci 59(7):1–12
Griffin G, Holub A, Perona P (2007) Caltech-256 object category dataset
He Y, Song K, Meng Q, Yan Y (2019) An end-to-end steel surface defect detection approach via fusing multiple hierarchical features. IEEE Trans Instrum Meas 69(4):1493–1504
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778
Hinton GE, Srivastava N, Krizhevsky A, Sutskever I, Salakhutdinov RR (2012) Improving neural networks by preventing co-adaptation of feature detectors. arXiv:1207.0580
Inoue H (2019) Multi-sample dropout for accelerated training and better generalization. arXiv:1905.09788
Kamili A, Fatima I, Hassan M, Parah SA, Vijaya Kumar V, Ambati L (2020) Embedding information reversibly in medical images for e-health. J Intell Fuzzy Syst 39(6):8389–8398
Khan N, Stavness I (2019) Sparseout: Controlling sparsity in deep networks. In: Canadian conference on artificial intelligence, Springer, pp 296–307
Konovalenko I, Maruschak P, Brezinová J, Viňáš J, Brezina J (2020) Steel surface defect classification using deep residual neural network. Metals 10(6):846
Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems 25
Lambert J, Sener O, Savarese S (2018) Deep learning under privileged information using heteroscedastic dropout. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 8886–8895
Li M, Xu G, Lai Q, Chen J (2022) A chaotic strategy-based quadratic opposition-based learning adaptive variable-speed whale optimization algorithm. Math Comput Simul 193:71–99
Lian Z, Jing X, Wang X, Huang H, Tan Y, Cui Y (2016) Dropconnect regularization method with sparsity constraint for neural networks. Chin J Electron 25(1):152–158
Molchanov D, Ashukha A, Vetrov D (2017) Variational dropout sparsifies deep neural networks. In: International conference on machine learning, PMLR, pp 2498–2507
Morerio P, Cavazza J, Volpi R, Vidal R, Murino V (2017) Curriculum dropout. In: Proceedings of the IEEE international conference on computer vision, pp 3544–3552
Nagaraj B, Arunkumar R, Nisi K, Vijayakumar P (2020) Enhancement of fraternal k-median algorithm with cnn for high dropout probabilities to evolve optimal time-complexity. Clust Comput 23(3):2001–2008
Ng ST, Skitmore M, Wong KF (2008) Using genetic algorithms and linear regression analysis for private housing demand forecast. Build Environ 43 (6):1171–1184
Nguyen S, Nguyen D, Nguyen K, Ho N, Than K, Bui H (2021) Improving bayesian inference in deep neural networks with variational structured dropout. arXiv e-prints 2102
Ou Y, Chen J, Chen W, Cheng C, Zhu Y, Xiao W, Lv H (2022) A quasi-distributed fiber magnetic field sensor based on frequency-shifted interferometry fiber cavity ringdown technique. Opt Laser Technol, 146. https://doi.org/10.1016/j.optlastec.2021.10760
Rennie SJ, Goel V, Thomas S (2014) Annealed dropout training of deep networks. In: 2014 IEEE spoken language technology workshop (SLT), IEEE, pp 159–164
Roccetti M, Delnevo G, Casini L, Cappiello G (2019) Is bigger always better? a controversial journey to the center of machine learning design, with uses and misuses of big data for predicting water meter failures. J Big Data 6(1):1–23
Sai Ambati L, El-Gayar OF, Nawar N (2020) Influence of the digital divide and socio-economic factors on prevalence of diabetes
Santra B, Paul A, Mukherjee DP (2020) Deterministic dropout for deep neural networks using composite random forest. Pattern Recogn Lett 131:205–212
Shen X, Tian X, Liu T, Xu F, Tao D (2017) Continuous dropout. IEEE Trans Neural Netw Learn Syst 29(9):3926–3937
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556
Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R (2014) Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res 15(1):1929–1958
Sun Y, Wang X, Tang X (2016) Sparsifying neural network connections for face recognition
Tang Y, Wang Y, Xu Y, Shi B, Xu C, Xu C, Xu C (2020) Beyond dropout: feature map distortion to regularize deep neural networks. In: Proceedings of the AAAI conference on artificial intelligence, vol 34, pp 5964–5971
Viloria A, Lezama OBP, Mercado-Caruzo N (2020) Unbalanced data processing using oversampling: machine learning. Procedia Comput Sci 175:108–113
Vinyals O, Blundell C, Lillicrap T, Wierstra D, et al. (2016) Matching networks for one shot learning. Advances in neural information processing systems 29
Wan L, Zeiler M, Zhang S, Le Cun Y, Fergus R (2013) Regularization of neural networks using dropconnect. In: International conference on machine learning, PMLR, pp 1058–1066
Wang G, Zhai Q (2021) Feature fusion network based on strip pooling. Sci Rep 11(1):1–8
Xie J, Ma Z, Lei J, Zhang G, Xue J-H, Tan Z-H, Guo J (2021) Advanced dropout: a model-free methodology for bayesian dropout optimization. IEEE Transactions on Pattern Analysis and Machine Intelligence
Yu F, Xu X (2014) A short-term load forecasting model of natural gas based on optimized genetic algorithm and improved bp neural network. Appl Energy 134:102–113
Zeng L, Shu W, Liu Z, Zou X, Wang S, Xia J, Xu C, Xiong D, Yang Z (2022) Vision-based high-precision intelligent monitoring for shield tail clearance. Autom Constr 134:104088
Zhou R, Guo F, Azarpazhooh MR, Spence JD, Ukwatta E, Ding M, Fenster A (2020) A voxel-based fully convolution network and continuous max-flow for carotid vessel-wall-volume segmentation from 3d ultrasound images. IEEE Trans Med Imaging 39(9):2844–2855
Zunino A, Bargal SA, Morerio P, Zhang J, Sclaroff S, Murino V (2021) Excitation dropout: encouraging plasticity in deep neural networks. Int J Comput Vis 129(4):1139–1152
Funding
This work was in part supported by the Key Research and Development Project of Hubei Province (No. 2020BAB114), the Key Project of Science and Technology Research Program of Hubei Educational Committee (No. D20211402), the Project of Xiangyang Industrial Institute of Hubei University of Technology (No. XYYJ2022C04), and the Open Foundation of Hubei Key Laboratory for High-efficiency Utilization of Solar Energy and Operation Control of Energy Storage System (No. HBSEES201903 & HBSEES202106).
Author information
Authors and Affiliations
Contributions
Liang Zeng, Hao Zhang, Yanyan Li, Maodong Li and Shanshan Wang conceived the experiments, Hao Zhang conducted the experiments. All authors reviewed the manuscript.
Corresponding authors
Ethics declarations
Ethics approval
This article does not contain any studies with human participants or animals performed by any of the authors.
Consent for Publication
Completed at Hubei University of Technology on February 21, 2022.
Conflict of Interests
The authors declare no competing interests.
Additional information
Consent to participate
Welcome readers to communicate.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Hao Zhang, Yanyan Li, Maodong Li and Shanshan Wang are contributed equally to this work.
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
Zeng, L., Zhang, H., Li, Y. et al. Supervision dropout: guidance learning in deep neural network. Multimed Tools Appl 82, 18831–18850 (2023). https://doi.org/10.1007/s11042-022-14274-0
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-022-14274-0