MLP 층을 갖는 CNN의 설계
Design of CNN with MLP Layer
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
After CNN basic structure was introduced by LeCun in 1989, there has not been a major structure change except for more deep network until recently. The deep network enhances the expression power due to improve the abstraction ability of the network, and can learn complex problems by increasing non linearity. However, the learning of a deep network means that it has vanishing gradient or longer learning time. In this study, we proposes a CNN structure with MLP layer. The proposed CNNs are superior to the general CNN in their classification performance. It is confirmed that classification accuracy is high due to include MLP layer which improves non linearity by experiment. In order to increase the performance without making a deep network, it is confirmed that the performance is improved by increasing the non linearity of the network.
Abstract
After CNN basic structure was introduced by LeCun in 1989, there has not been a major structure change except for more deep network until recently. The deep network enhances the expression power due to improve the abstraction ability of the network, and can learn complex problems by increasing non linearity. However, the learning of a deep network means that it has vanishing gradient or longer learning time. In this study, we proposes a CNN structure with MLP layer. The proposed CNNs are superior to the general CNN in their classification performance. It is confirmed that classification accuracy is high due to include MLP layer which improves non linearity by experiment. In order to increase the performance without making a deep network, it is confirmed that the performance is improved by increasing the non linearity of the network.
참고문헌 (12)
- Guang Chen, "Automatic Fish Classification System Using Deep Learning", International Conference on Tools with Artificial Intelligence (2017), IEEE computer society
- J. Cao, "Pedestrian detection inspired by appearance constancy and shape symmetry", Proc. IEEE Int. Conf. Comput. Vis. Pattern Recognit, pp.1316~1324 (2016)
- 이태환, "신경회로망을 사용한 증기표 전체 영역의 모델링", 한국기계기술학회지, 20(1), pp.82~88 (2018), 한국기계기술학회[DOI]
- Y. LeCun, "Gradient-based learning applied to document recognition", Proceedings of the IEEE, 86(11), pp.2278~2324 (1998)
- Y. Bengio, "Learning deep architectures for AI", Foundations and Trends in Machine Learning, 2(1), pp.1~127 (2009)
- G. E. Hinton, "A fast learning algorithm for deep belief nets", Neural Computation, 18(7), pp.1527~1554 (2006)
- 이언호, "3차원 점군 데이터를 이용한 수정 구형 특징 표현기 개발 및 도시 구조물 분류를 위한 컨볼루션 신경망의 응용", 한국기계기술학회지, 20(1), pp.18~25 (2018), 한국기계기술학회[DOI]
- Alex Krizhevsky, "Imagenet classification with deep convolutional neural networks", Advances in neural information processing systems, 25, pp.1~9 (2012)
- G. E. Hinton, "Reducing the dimensionality of data with neural networks", Science, 313, pp.504~507 (2006)
- "Network in network" (2013)[URL]
- Yanwei Pang, "Convolution in Convolution for Network in Network", IEEE Trans. on Neural Networks and Learning Systems, 29(5) (2018)