Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks
With the success of deep learning in recent years, lots of different AI models have been applied to the real world. At the same time, how to train a model with good performance becomes a problem people have to face. One of the most important things is hyperparameter tuning since it determines the se...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9987510/ |
| _version_ | 1811169681773428736 |
|---|---|
| author | Chi-Lin Hsieh Kuei-Chung Chang Tien-Fu Chen |
| author_facet | Chi-Lin Hsieh Kuei-Chung Chang Tien-Fu Chen |
| author_sort | Chi-Lin Hsieh |
| collection | DOAJ |
| description | With the success of deep learning in recent years, lots of different AI models have been applied to the real world. At the same time, how to train a model with good performance becomes a problem people have to face. One of the most important things is hyperparameter tuning since it determines the setting of training flow. However, most of the conventional hyperparameter algorithms are inefficient, because they usually search from scratch for each new task, and that’s why they require large search trials to find a good combination of hyperparameters. In this paper, we present a systematic hyperparameter tuning framework which utilizes prior knowledge with a suggestion algorithm and an adaptive controller to improve its efficiency rather than search from scratch for each task. In this way, our proposed method can achieve a better performance with the same budget. In the experiments, we applied our methods to tens of popular datasets, and the results show that our proposed methods can outperform than other approaches. |
| first_indexed | 2024-04-10T16:46:16Z |
| format | Article |
| id | doaj.art-284b8a9c294a48b9b69cc9693ebaef40 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T16:46:16Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-284b8a9c294a48b9b69cc9693ebaef402023-02-08T00:01:05ZengIEEEIEEE Access2169-35362023-01-0111110891110110.1109/ACCESS.2022.32293909987510Adaptive Similarity-Aware Hyperparameter Tuners for Classification TasksChi-Lin Hsieh0Kuei-Chung Chang1https://orcid.org/0000-0003-4145-1927Tien-Fu Chen2Department of Computer Science, National Yang Ming Chiao Tung University, Hsinchu, TaiwanInternational Graduate School of Artificial Intelligence, National Yunlin University of Science and Technology, Yunlin, TaiwanDepartment of Computer Science, National Yang Ming Chiao Tung University, Hsinchu, TaiwanWith the success of deep learning in recent years, lots of different AI models have been applied to the real world. At the same time, how to train a model with good performance becomes a problem people have to face. One of the most important things is hyperparameter tuning since it determines the setting of training flow. However, most of the conventional hyperparameter algorithms are inefficient, because they usually search from scratch for each new task, and that’s why they require large search trials to find a good combination of hyperparameters. In this paper, we present a systematic hyperparameter tuning framework which utilizes prior knowledge with a suggestion algorithm and an adaptive controller to improve its efficiency rather than search from scratch for each task. In this way, our proposed method can achieve a better performance with the same budget. In the experiments, we applied our methods to tens of popular datasets, and the results show that our proposed methods can outperform than other approaches.https://ieeexplore.ieee.org/document/9987510/Classificationdataset similarityevolutionary algorithmhyperparameter tuningtransfer learning |
| spellingShingle | Chi-Lin Hsieh Kuei-Chung Chang Tien-Fu Chen Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks IEEE Access Classification dataset similarity evolutionary algorithm hyperparameter tuning transfer learning |
| title | Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks |
| title_full | Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks |
| title_fullStr | Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks |
| title_full_unstemmed | Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks |
| title_short | Adaptive Similarity-Aware Hyperparameter Tuners for Classification Tasks |
| title_sort | adaptive similarity aware hyperparameter tuners for classification tasks |
| topic | Classification dataset similarity evolutionary algorithm hyperparameter tuning transfer learning |
| url | https://ieeexplore.ieee.org/document/9987510/ |
| work_keys_str_mv | AT chilinhsieh adaptivesimilarityawarehyperparametertunersforclassificationtasks AT kueichungchang adaptivesimilarityawarehyperparametertunersforclassificationtasks AT tienfuchen adaptivesimilarityawarehyperparametertunersforclassificationtasks |