Comparison of semi-supervised learning algorithms
In this report, we conducted image classification experiments in a semi-supervised setting using three datasets of various sizes and content, CIFAR10, CIFAR100 and EuroSAT, with only 1000 samples labelled and the remaining as unlabelled for training. The performance and training duration of three SS...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project (FYP) |
| Language: | English |
| Published: |
Nanyang Technological University
2023
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/172003 |
| _version_ | 1826125221157928960 |
|---|---|
| author | Teo, Sheng Huai |
| author2 | Li Boyang |
| author_facet | Li Boyang Teo, Sheng Huai |
| author_sort | Teo, Sheng Huai |
| collection | NTU |
| description | In this report, we conducted image classification experiments in a semi-supervised setting using three datasets of various sizes and content, CIFAR10, CIFAR100 and EuroSAT, with only 1000 samples labelled and the remaining as unlabelled for training. The performance and training duration of three SSL algorithms, MixMatch, FixMatch and FlexMatch, were compared. For CIFAR10 and EuroSAT, MixMatch achieved the highest accuracy of 95.8% and 93.4% respectively. Despite having the best performance for both datasets, MixMatch took most time to train, with an average of 27.5% longer than FixMatch, which has the shortest training duration. For CIFAR100, FixMatch obtained the best results for all four metrics obtained. FlexMatch was the next best, with an accuracy of 73.2%, and the other metrics having a difference of roughly 2.5% compared to FixMatch. |
| first_indexed | 2024-10-01T06:33:09Z |
| format | Final Year Project (FYP) |
| id | ntu-10356/172003 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:33:09Z |
| publishDate | 2023 |
| publisher | Nanyang Technological University |
| record_format | dspace |
| spelling | ntu-10356/1720032023-11-24T15:37:39Z Comparison of semi-supervised learning algorithms Teo, Sheng Huai Li Boyang School of Computer Science and Engineering boyang.li@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence In this report, we conducted image classification experiments in a semi-supervised setting using three datasets of various sizes and content, CIFAR10, CIFAR100 and EuroSAT, with only 1000 samples labelled and the remaining as unlabelled for training. The performance and training duration of three SSL algorithms, MixMatch, FixMatch and FlexMatch, were compared. For CIFAR10 and EuroSAT, MixMatch achieved the highest accuracy of 95.8% and 93.4% respectively. Despite having the best performance for both datasets, MixMatch took most time to train, with an average of 27.5% longer than FixMatch, which has the shortest training duration. For CIFAR100, FixMatch obtained the best results for all four metrics obtained. FlexMatch was the next best, with an accuracy of 73.2%, and the other metrics having a difference of roughly 2.5% compared to FixMatch. Bachelor of Engineering (Computer Science) 2023-11-20T06:21:50Z 2023-11-20T06:21:50Z 2023 Final Year Project (FYP) Teo, S. H. (2023). Comparison of semi-supervised learning algorithms. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/172003 https://hdl.handle.net/10356/172003 en SCSE22-0770 application/pdf Nanyang Technological University |
| spellingShingle | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Teo, Sheng Huai Comparison of semi-supervised learning algorithms |
| title | Comparison of semi-supervised learning algorithms |
| title_full | Comparison of semi-supervised learning algorithms |
| title_fullStr | Comparison of semi-supervised learning algorithms |
| title_full_unstemmed | Comparison of semi-supervised learning algorithms |
| title_short | Comparison of semi-supervised learning algorithms |
| title_sort | comparison of semi supervised learning algorithms |
| topic | Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence |
| url | https://hdl.handle.net/10356/172003 |
| work_keys_str_mv | AT teoshenghuai comparisonofsemisupervisedlearningalgorithms |