Network Pathway Extraction Focusing on Object Level
In this paper, I propose an efficient method of identifying important neurons that are related to an object’s concepts by mainly considering the relationship between these neurons and their object concept or class. I first quantify the activation values among neurons, based on which histograms of ea...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Eng |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-4117/4/1/9 |
| _version_ | 1797612039676362752 |
|---|---|
| author | Ali Alqahtani |
| author_facet | Ali Alqahtani |
| author_sort | Ali Alqahtani |
| collection | DOAJ |
| description | In this paper, I propose an efficient method of identifying important neurons that are related to an object’s concepts by mainly considering the relationship between these neurons and their object concept or class. I first quantify the activation values among neurons, based on which histograms of each neuron are generated. Then, the obtained histograms are clustered to identify the neurons’ importance. A network-wide holistic approach is also introduced to efficiently identify important neurons and their influential connections to reveal the pathway of a given class. The influential connections as well as their important neurons are carefully evaluated to reveal the sub-network of each object’s concepts. The experimental results on the MNIST and Fashion MNIST datasets show the effectiveness of the proposed method. |
| first_indexed | 2024-03-11T06:36:21Z |
| format | Article |
| id | doaj.art-e6bda9f3d21a45938df83dac92fe7288 |
| institution | Directory Open Access Journal |
| issn | 2673-4117 |
| language | English |
| last_indexed | 2024-03-11T06:36:21Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Eng |
| spelling | doaj.art-e6bda9f3d21a45938df83dac92fe72882023-11-17T10:53:06ZengMDPI AGEng2673-41172023-01-014115115810.3390/eng4010009Network Pathway Extraction Focusing on Object LevelAli Alqahtani0Department of Computer Science, King Khalid University, Abha 61421, Saudi ArabiaIn this paper, I propose an efficient method of identifying important neurons that are related to an object’s concepts by mainly considering the relationship between these neurons and their object concept or class. I first quantify the activation values among neurons, based on which histograms of each neuron are generated. Then, the obtained histograms are clustered to identify the neurons’ importance. A network-wide holistic approach is also introduced to efficiently identify important neurons and their influential connections to reveal the pathway of a given class. The influential connections as well as their important neurons are carefully evaluated to reveal the sub-network of each object’s concepts. The experimental results on the MNIST and Fashion MNIST datasets show the effectiveness of the proposed method.https://www.mdpi.com/2673-4117/4/1/9deep learningnetwork pathway extractionneuron importance |
| spellingShingle | Ali Alqahtani Network Pathway Extraction Focusing on Object Level Eng deep learning network pathway extraction neuron importance |
| title | Network Pathway Extraction Focusing on Object Level |
| title_full | Network Pathway Extraction Focusing on Object Level |
| title_fullStr | Network Pathway Extraction Focusing on Object Level |
| title_full_unstemmed | Network Pathway Extraction Focusing on Object Level |
| title_short | Network Pathway Extraction Focusing on Object Level |
| title_sort | network pathway extraction focusing on object level |
| topic | deep learning network pathway extraction neuron importance |
| url | https://www.mdpi.com/2673-4117/4/1/9 |
| work_keys_str_mv | AT alialqahtani networkpathwayextractionfocusingonobjectlevel |