{"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/learning-to-predict-by-the-methods-of","title":"Learning to Predict by the Methods of Temporal Differences","arxiv_id":null,"date":"1988-03-09","proceeding":"Machine Learning 1988 3","authors":["Richard S. Sutton"],"abstract":"This article introduces a class of incremental learning procedures specialized for prediction that is, for using past experience with an incompletely known\r\nsystem to predict its future behavior. Whereas conventional prediction-learning\r\nmethods assign credit by means of the difference between predicted and actual outcomes, tile new methods assign credit by means of the difference between temporally\r\nsuccessive predictions. Although such temporal-difference method~ have been used in\r\nSamuel's checker player, Holland's bucket brigade, and the author's Adaptive Heuristic Critic, they have remained poorly understood. Here we prove their convergence\r\nand optimality for special cases and relate them to supervised-learning methods. For\r\nmost real-world prediction problems, telnporal-differenee methods require less memory and less peak computation than conventional methods and they produce more\r\naccurate predictions. We argue that most problems to which supervised learning\r\nis currently applied are really prediction problems of the sort to which temporaldifference methods can be applied to advantage.","url_abs":"http://incompleteideas.net/papers/sutton-88-with-erratum.pdf","url_pdf":"http://incompleteideas.net/papers/sutton-88-with-erratum.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":"learning-to-predict-by-the-methods-of","repo_url":"https://github.com/mindspore-courses/Rainbow-MindSpore","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"incremental-learning","task_name":"Incremental Learning"},{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}