EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks
This study aims to develop a robust metalearning system for rapid classification on a large number of tasks. The model-agnostic metalearning (MAML) with the CACTUs method (clustering to automatically construct tasks for unsupervised metalearning) is improved as EW-CACTUs-MAML after integrated with t...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2022-01-01
|
| Series: | Complexity |
| Online Access: | http://dx.doi.org/10.1155/2022/7330823 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832567762379079680 |
|---|---|
| author | Wen-Feng Wang Jingjing Zhang Peng An |
| author_facet | Wen-Feng Wang Jingjing Zhang Peng An |
| author_sort | Wen-Feng Wang |
| collection | DOAJ |
| description | This study aims to develop a robust metalearning system for rapid classification on a large number of tasks. The model-agnostic metalearning (MAML) with the CACTUs method (clustering to automatically construct tasks for unsupervised metalearning) is improved as EW-CACTUs-MAML after integrated with the entropy weight (EW) method. Few-shot mechanisms are introduced in the deep network for efficient learning of a large number of tasks. The process of implementation is theoretically interpreted as “gene intelligence.” Validation of EW-CACTUs-MAML on a typical dataset (Omniglot) indicates an accuracy of 97.42%, performing better than CACTUs-MAML (validation accuracy = 97.22%). At the end of this paper, the availability of our thoughts to improve another metalearning system (EW-CACTUs-ProtoNets) is also preliminarily discussed based on a cross-validation on another typical dataset (Miniimagenet). |
| format | Article |
| id | doaj-art-dcb5d1e4193c4d62a0efc4ed1bbf4fff |
| institution | Kabale University |
| issn | 1099-0526 |
| language | English |
| publishDate | 2022-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Complexity |
| spelling | doaj-art-dcb5d1e4193c4d62a0efc4ed1bbf4fff2025-02-03T01:00:46ZengWileyComplexity1099-05262022-01-01202210.1155/2022/7330823EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of TasksWen-Feng Wang0Jingjing Zhang1Peng An2Shanghai Institute of TechnologyShanghai Institute of TechnologyNingbo University of TechnologyThis study aims to develop a robust metalearning system for rapid classification on a large number of tasks. The model-agnostic metalearning (MAML) with the CACTUs method (clustering to automatically construct tasks for unsupervised metalearning) is improved as EW-CACTUs-MAML after integrated with the entropy weight (EW) method. Few-shot mechanisms are introduced in the deep network for efficient learning of a large number of tasks. The process of implementation is theoretically interpreted as “gene intelligence.” Validation of EW-CACTUs-MAML on a typical dataset (Omniglot) indicates an accuracy of 97.42%, performing better than CACTUs-MAML (validation accuracy = 97.22%). At the end of this paper, the availability of our thoughts to improve another metalearning system (EW-CACTUs-ProtoNets) is also preliminarily discussed based on a cross-validation on another typical dataset (Miniimagenet).http://dx.doi.org/10.1155/2022/7330823 |
| spellingShingle | Wen-Feng Wang Jingjing Zhang Peng An EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks Complexity |
| title | EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks |
| title_full | EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks |
| title_fullStr | EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks |
| title_full_unstemmed | EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks |
| title_short | EW-CACTUs-MAML: A Robust Metalearning System for Rapid Classification on a Large Number of Tasks |
| title_sort | ew cactus maml a robust metalearning system for rapid classification on a large number of tasks |
| url | http://dx.doi.org/10.1155/2022/7330823 |
| work_keys_str_mv | AT wenfengwang ewcactusmamlarobustmetalearningsystemforrapidclassificationonalargenumberoftasks AT jingjingzhang ewcactusmamlarobustmetalearningsystemforrapidclassificationonalargenumberoftasks AT pengan ewcactusmamlarobustmetalearningsystemforrapidclassificationonalargenumberoftasks |