Modular learning of convolutional neural networks
In recent years, a wide variety of network structures and training methods have been proposed for deep learning. However, the underlying mechanism of a deep network is not fully understood, which is considered as a black box system. The layer-wise learning method was proposed a decade ago, but now i...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis-Master by Coursework |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/163999 |
| _version_ | 1826117295899934720 |
|---|---|
| author | Wang, Jinhua |
| author2 | Cheah Chien Chern |
| author_facet | Cheah Chien Chern Wang, Jinhua |
| author_sort | Wang, Jinhua |
| collection | NTU |
| description | In recent years, a wide variety of network structures and training methods have been proposed for deep learning. However, the underlying mechanism of a deep network is not fully understood, which is considered as a black box system. The layer-wise learning method was proposed a decade ago, but now it is rarely used due to the trade-off in performance as compared to the standard end-to-end learning. Recently, layer-wise learning has been considered for application in interpretable or analytical neural networks. Therefore, a key target is to improve the performance of the layer-wise learning.
In this dissertation, a modular deep learning method is developed on the basis of classical layer-wise learning. In addition, the network performance is further improved by proposing epoch-wise learning on the basis of modular deep learning. Through the case studies using several common datasets, the proposed approaches are compared with the traditional layer-wise learning method in terms of performance. |
| first_indexed | 2024-10-01T04:25:17Z |
| format | Thesis-Master by Coursework |
| id | ntu-10356/163999 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:25:17Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1639992023-01-03T05:47:01Z Modular learning of convolutional neural networks Wang, Jinhua Cheah Chien Chern School of Electrical and Electronic Engineering ECCCheah@ntu.edu.sg Engineering::Electrical and electronic engineering Engineering::Computer science and engineering In recent years, a wide variety of network structures and training methods have been proposed for deep learning. However, the underlying mechanism of a deep network is not fully understood, which is considered as a black box system. The layer-wise learning method was proposed a decade ago, but now it is rarely used due to the trade-off in performance as compared to the standard end-to-end learning. Recently, layer-wise learning has been considered for application in interpretable or analytical neural networks. Therefore, a key target is to improve the performance of the layer-wise learning. In this dissertation, a modular deep learning method is developed on the basis of classical layer-wise learning. In addition, the network performance is further improved by proposing epoch-wise learning on the basis of modular deep learning. Through the case studies using several common datasets, the proposed approaches are compared with the traditional layer-wise learning method in terms of performance. Master of Science (Computer Control and Automation) 2023-01-03T05:47:01Z 2023-01-03T05:47:01Z 2022 Thesis-Master by Coursework Wang, J. (2022). Modular learning of convolutional neural networks. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/163999 https://hdl.handle.net/10356/163999 en ISM-DISS-03097 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Electrical and electronic engineering Engineering::Computer science and engineering Wang, Jinhua Modular learning of convolutional neural networks |
| title | Modular learning of convolutional neural networks |
| title_full | Modular learning of convolutional neural networks |
| title_fullStr | Modular learning of convolutional neural networks |
| title_full_unstemmed | Modular learning of convolutional neural networks |
| title_short | Modular learning of convolutional neural networks |
| title_sort | modular learning of convolutional neural networks |
| topic | Engineering::Electrical and electronic engineering Engineering::Computer science and engineering |
| url | https://hdl.handle.net/10356/163999 |
| work_keys_str_mv | AT wangjinhua modularlearningofconvolutionalneuralnetworks |