{"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/multitask-learning-a-knowledge-based-source","title":"Multitask Learning: A Knowledge-Based Source of Inductive Bias","arxiv_id":null,"date":"1993-01-01","proceeding":"ICML 1993 1","authors":["Richard A. Caruana"],"abstract":"This paper suggests that it may be easier to learn\r\nseveral hard tasks at one time than to learn these\r\nsame tasks separately. In effect, the information provided by the training signal for each task\r\nserves as a domain-specific inductive bias for the\r\nother tasks. Frequently the world gives us clusters\r\nof related tasks to learn. When it does not, it is often straightforward to create additional tasks. For\r\nmany domains, acquiring inductive bias by collecting additional teaching signal may be more\r\npractical than the traditional approach of codifying domain-specific biases acquired from human\r\nexpertise. We call this approach Multitask Learning (MTL). Since much of the power of an inductive learner follows directly from its inductive\r\nbias, multitask learning may yield more powerful learning. An empirical example of multitask\r\nconnectionist learning is presented where learning improves by training one network on several\r\nrelated tasks at the same time. Multitask decision\r\ntree induction is also outlined.","url_abs":"https://www.semanticscholar.org/paper/Multitask-Learning%3A-A-Knowledge-Based-Source-of-Caruana/9464d15f4f8d578f93332db4aa1c9c182fd51735","url_pdf":"http://citeseerx.ist.psu.edu/viewdoc/download;jsessionid=F45CA386897E5A6EBCF74D5DBAC85A13?doi=10.1.1.57.3196&rep=rep1&type=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":"multitask-learning-a-knowledge-based-source","repo_url":"https://github.com/PaddlePaddle/PaddleRec/tree/release/2.1.0/models/multitask/share_bottom","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"multitask-learning-a-knowledge-based-source","repo_url":"https://github.com/shenweichen/DeepCTR","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"inductive-bias","task_name":"Inductive Bias"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}