Composing Diverse Ensembles of Convolutional Neural Networks by Penalization

Author:

Harangi Balazs1ORCID,Baran Agnes1,Beregi-Kovacs Marcell1ORCID,Hajdu Andras1ORCID

Affiliation:

1. Faculty of Informatics, University of Debrecen, H-4028 Debrecen, Hungary

Abstract

Ensemble-based systems are well known to have the capacity to outperform individual approaches if the ensemble members are sufficiently accurate and diverse. This paper investigates how an efficient ensemble of deep convolutional neural networks (CNNs) can be created by forcing them to adjust their parameters during the training process to increase diversity in their decisions. As a new theoretical approach to reach this aim, we join the member neural architectures via a fully connected layer and insert a new correlation penalty term in the loss function to obstruct their similar operation. With this complementary term, we implement the standard guideline of ensemble creation to increase the members’ diversity for CNNs in a more detailed and flexible way than similar existing techniques. As for applicability, we show that our approach can be efficiently used in various classification tasks. More specifically, we demonstrate its performance in challenging medical image analysis and natural image classification problems. Besides the theoretical considerations and foundations, our experimental findings suggest that the proposed technique is competitive. Namely, on the one hand, the classification rate of the ensemble trained in this way outperformed all the individual accuracies of the state-of-the-art member CNNs according to the standard error functions of these application domains. On the other hand, it is also validated that the ensemble members get more diverse and their accuracies are raised by adding the penalization term. Moreover, we performed a full comparative analysis, including other state-of-the-art ensemble-based approaches recommended for the same classification tasks. This comparative study also confirmed the superiority of our method, as it overcame the current solutions.

Funder

New National Excellence Program of the Ministry for Innovation and Technology from the source of the National Research, Development and Innovation Fund

Janos Bolyai Research Scholarship of the Hungarian Academy of Sciences

Publisher

MDPI AG

Subject

General Mathematics,Engineering (miscellaneous),Computer Science (miscellaneous)

Reference56 articles.

1. Zhang, Y., Sohn, K., Villegas, R., Pan, G., and Lee, H. (2015, January 7–12). Improving object detection with deep convolutional networks via Bayesian optimization and structured prediction. Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA.

2. Zhang, D., Javed, O., and Shah, M. (2013, January 23–28). Video Object Segmentation through Spatially Accurate and Temporally Dense Extraction of Primary Object Regions. Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA.

3. Sermanet, P., Eigen, D., Zhang, X., Mathieu, M., Fergus, R., and LeCun, Y. (2014, January 14–16). OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks. Proceedings of the 2nd International Conference on Learning Representations, Banff, AB, Canada. Available online: http://arxiv.org/abs/1312.6229.

4. Girshick, R., Donahue, J., Darrell, T., and Malik, J. (2014, January 23–28). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.

5. Fleet, D., Pajdla, T., Schiele, B., and Tuytelaars, T. (2014, January 23–28). Part-Based R-CNNs for Fine-Grained Category Detection. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Advanced Multi-Label Image Classification Techniques Using Ensemble Methods;Machine Learning and Knowledge Extraction;2024-06-07

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3