{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/easy-transfer-learning-by-exploiting-intra","title":"Easy Transfer Learning By Exploiting Intra-domain Structures","arxiv_id":"1904.01376","date":"2019-04-02","proceeding":null,"authors":["Jindong Wang","Yiqiang Chen","Han Yu","Meiyu Huang","Qiang Yang"],"abstract":"Transfer learning aims at transferring knowledge from a well-labeled domain\nto a similar but different domain with limited or no labels. Unfortunately,\nexisting learning-based methods often involve intensive model selection and\nhyperparameter tuning to obtain good results. Moreover, cross-validation is not\npossible for tuning hyperparameters since there are often no labels in the\ntarget domain. This would restrict wide applicability of transfer learning\nespecially in computationally-constraint devices such as wearables. In this\npaper, we propose a practically Easy Transfer Learning (EasyTL) approach which\nrequires no model selection and hyperparameter tuning, while achieving\ncompetitive performance. By exploiting intra-domain structures, EasyTL is able\nto learn both non-parametric transfer features and classifiers. Extensive\nexperiments demonstrate that, compared to state-of-the-art traditional and deep\nmethods, EasyTL satisfies the Occam's Razor principle: it is extremely easy to\nimplement and use while achieving comparable or better performance in\nclassification accuracy and much better computational efficiency. Additionally,\nit is shown that EasyTL can increase the performance of existing transfer\nfeature learning methods.","url_abs":"http://arxiv.org/abs/1904.01376v2","url_pdf":"http://arxiv.org/pdf/1904.01376v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"easy-transfer-learning-by-exploiting-intra","repo_url":"https://github.com/jindongwang/transferlearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"model-selection","task_name":"Model Selection"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/domain-adaptation-on-imageclef-da","task":"Domain Adaptation","dataset":"ImageCLEF-DA","model":"EasyTL","rank_in_archive_order":13,"of":17,"metrics":{"Accuracy":"88.2"},"uses_additional_data":false},{"leaderboard":"/sota/transfer-learning-on-office-home","task":"Transfer Learning","dataset":"Office-Home","model":"EasyTL","rank_in_archive_order":4,"of":5,"metrics":{"Accuracy":"63.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.01376","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}