Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth
This study originates a new model, the Feature Parallelism Model (FPM), and compares it to deep learning models along depth, which is the number of layers that comprises a machine learning model. It is the number of layers in the horizontal axis, in the case of FPM. We found that only 6 layers optim...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AESS Publication
2019
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/75336/ http://irep.iium.edu.my/75336/1/NEUROSCIENCE-INSPIRED%20ARTIFICIAL%20VISION%20FEATURE.pdf |
| _version_ | 1848788108375490560 |
|---|---|
| author | Hassan, Marwa Yousif Khalifa, Othman Omran Hassan Abdalla Hashim, Aisha |
| author_facet | Hassan, Marwa Yousif Khalifa, Othman Omran Hassan Abdalla Hashim, Aisha |
| author_sort | Hassan, Marwa Yousif |
| building | IIUM Repository |
| collection | Online Access |
| description | This study originates a new model, the Feature Parallelism Model (FPM), and compares it to deep learning models along depth, which is the number of layers that comprises a machine learning model. It is the number of layers in the horizontal axis, in the case of FPM. We found that only 6 layers optimize FPM‟s performance. FPM has been inspired by the human brain and follows some organizing principles that underlie the human visual system. We review here the standard practice in deep learning, which is opting in to the deepest model that the computational resources allow up to hundreds
of layers, seeking better accuracies. We have implemented FPM using 5, 6, 7, and 8
layers and observed accuracy as well as training time for each. We show that much less
depth is needed for FPM, down to 6 layers. This optimizes both accuracy and training
time for the model. Moreover, in a previous study we have proposed the model and
have shown that while FPM uses less computational resources proved by 21%
reduction in training time, it performs as well as deep learning regarding models‟
accuracy. |
| first_indexed | 2025-11-14T17:35:35Z |
| format | Article |
| id | iium-75336 |
| institution | International Islamic University Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-14T17:35:35Z |
| publishDate | 2019 |
| publisher | AESS Publication |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | iium-753362019-12-06T01:28:49Z http://irep.iium.edu.my/75336/ Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth Hassan, Marwa Yousif Khalifa, Othman Omran Hassan Abdalla Hashim, Aisha TK5101 Telecommunication. Including telegraphy, radio, radar, television This study originates a new model, the Feature Parallelism Model (FPM), and compares it to deep learning models along depth, which is the number of layers that comprises a machine learning model. It is the number of layers in the horizontal axis, in the case of FPM. We found that only 6 layers optimize FPM‟s performance. FPM has been inspired by the human brain and follows some organizing principles that underlie the human visual system. We review here the standard practice in deep learning, which is opting in to the deepest model that the computational resources allow up to hundreds of layers, seeking better accuracies. We have implemented FPM using 5, 6, 7, and 8 layers and observed accuracy as well as training time for each. We show that much less depth is needed for FPM, down to 6 layers. This optimizes both accuracy and training time for the model. Moreover, in a previous study we have proposed the model and have shown that while FPM uses less computational resources proved by 21% reduction in training time, it performs as well as deep learning regarding models‟ accuracy. AESS Publication 2019 Article PeerReviewed application/pdf en http://irep.iium.edu.my/75336/1/NEUROSCIENCE-INSPIRED%20ARTIFICIAL%20VISION%20FEATURE.pdf Hassan, Marwa Yousif and Khalifa, Othman Omran and Hassan Abdalla Hashim, Aisha (2019) Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth. Journal of Asian Scientific Research, 9 (9). pp. 127-139. ISSN 2226-5724 E-ISSN 2223-1331 http://www.aessweb.com/pdf-files/JASR-2019-9(9)-127-139.pdf 10.18488/journal.2.2019.99.127.139 |
| spellingShingle | TK5101 Telecommunication. Including telegraphy, radio, radar, television Hassan, Marwa Yousif Khalifa, Othman Omran Hassan Abdalla Hashim, Aisha Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title | Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title_full | Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title_fullStr | Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title_full_unstemmed | Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title_short | Neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| title_sort | neuroscience-inspired artificial vision feature parallelism and deep learning models, a comparative study ii depth |
| topic | TK5101 Telecommunication. Including telegraphy, radio, radar, television |
| url | http://irep.iium.edu.my/75336/ http://irep.iium.edu.my/75336/ http://irep.iium.edu.my/75336/ http://irep.iium.edu.my/75336/1/NEUROSCIENCE-INSPIRED%20ARTIFICIAL%20VISION%20FEATURE.pdf |