{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,18]],"date-time":"2026-01-18T10:15:11Z","timestamp":1768731311485,"version":"3.49.0"},"reference-count":54,"publisher":"SAGE Publications","issue":"2","license":[{"start":{"date-parts":[[2014,2,7]],"date-time":"2014-02-07T00:00:00Z","timestamp":1391731200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Adaptive Behavior"],"published-print":{"date-parts":[[2014,4]]},"abstract":"<jats:p> The term \u2018nexting\u2019 has been used by psychologists to refer to the propensity of people and many other animals to continually predict what will happen next in an immediate, local, and personal sense. The ability to \u2018next\u2019 constitutes a basic kind of awareness and knowledge of one\u2019s environment. In this paper we present results with a robot that learns to next in real time, making thousands of predictions about sensory input signals at timescales from 0.1 to 8 seconds. Our predictions are formulated as a generalization of the value functions commonly used in reinforcement learning, where now an arbitrary function of the sensory input signals is used as a pseudo reward, and the discount rate determines the timescale. We show that six thousand predictions, each computed as a function of six thousand features of the state, can be learned and updated online ten times per second on a laptop computer, using the standard temporal-difference( \u03bb) algorithm with linear function approximation. This approach is sufficiently computationally efficient to be used for real-time learning on the robot and sufficiently data efficient to achieve substantial accuracy within 30 minutes. Moreover, a single tile-coded feature representation suffices to accurately predict many different signals over a significant range of timescales. We also extend nexting beyond simple timescales by letting the discount rate be a function of the state and show that nexting predictions of this more general form can also be learned with substantial accuracy. General nexting provides a simple yet powerful mechanism for a robot to acquire predictive knowledge of the dynamics of its environment. <\/jats:p>","DOI":"10.1177\/1059712313511648","type":"journal-article","created":{"date-parts":[[2014,2,8]],"date-time":"2014-02-08T06:01:25Z","timestamp":1391839285000},"page":"146-160","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":29,"title":["Multi-timescale nexting in a reinforcement learning robot"],"prefix":"10.1177","volume":"22","author":[{"given":"Joseph","family":"Modayil","sequence":"first","affiliation":[{"name":"Reinforcement Learning and Artificial Intelligence Laboratory, University of Alberta, Canada"}]},{"given":"Adam","family":"White","sequence":"additional","affiliation":[{"name":"Reinforcement Learning and Artificial Intelligence Laboratory, University of Alberta, Canada"}]},{"given":"Richard S","family":"Sutton","sequence":"additional","affiliation":[{"name":"Reinforcement Learning and Artificial Intelligence Laboratory, University of Alberta, Canada"}]}],"member":"179","published-online":{"date-parts":[[2014,2,7]]},"reference":[{"key":"bibr1-1059712313511648","first-page":"396","volume-title":"Computer models of thought and language","author":"Becker J. D.","year":"1973"},{"key":"bibr2-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1177\/0278364911404092"},{"key":"bibr3-1059712313511648","volume-title":"Time series analysis: Forecasting and control","author":"Box G. E.","year":"2011"},{"key":"bibr4-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1037\/h0058944"},{"key":"bibr5-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/b11711"},{"key":"bibr6-1059712313511648","volume-title":"Model predictive control","author":"Camacho E. F.","year":"2004"},{"key":"bibr7-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1162\/089892900562318"},{"key":"bibr8-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X12000477"},{"key":"bibr9-1059712313511648","volume-title":"Intelligence: Its organization and development","author":"Cunningham M.","year":"1972"},{"key":"bibr10-1059712313511648","first-page":"271","volume":"5","author":"Dayan P.","year":"1993","journal-title":"Advances in Neural Information Processing Systems"},{"key":"bibr11-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1109\/ACC.2012.6315022"},{"key":"bibr12-1059712313511648","doi-asserted-by":"crossref","DOI":"10.7551\/mitpress\/4378.001.0001","volume-title":"Made-up minds: A constructivist approach to artificial intelligence","author":"Drescher G. L.","year":"1991"},{"key":"bibr13-1059712313511648","volume-title":"Stumbling on happiness","author":"Gilbert D.","year":"2006"},{"key":"bibr14-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1017\/S0140525X04000093"},{"key":"bibr15-1059712313511648","volume-title":"On intelligence","author":"Hawkins J.","year":"2004"},{"key":"bibr16-1059712313511648","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/6575.001.0001"},{"key":"bibr17-1059712313511648","first-page":"1094","volume-title":"Proceedings of International Joint Conference on Artificial Intelligence","author":"Kaelbling L.","year":"1993"},{"key":"bibr18-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511546877"},{"key":"bibr19-1059712313511648","volume-title":"This is your brain on music","author":"Levitin D.","year":"2006"},{"key":"bibr20-1059712313511648","first-page":"1555","volume":"14","author":"Littman M. L.","year":"2002","journal-title":"Advances in Neural Information Processing Systems"},{"key":"bibr21-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4612-1768-8_11"},{"key":"bibr22-1059712313511648","unstructured":"Maei H. R. (2011). Gradient temporal-difference learning algorithms. PhD Thesis, University of Alberta, Canada."},{"key":"bibr23-1059712313511648","doi-asserted-by":"publisher","DOI":"10.2991\/agi.2010.22"},{"key":"bibr24-1059712313511648","first-page":"A1","author":"Markoff J.","year":"2010","journal-title":"The New York Times"},{"key":"bibr25-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33093-3_30"},{"key":"bibr26-1059712313511648","first-page":"846","volume-title":"Proceedings of the Seventeenth Conference of the Association for the Advancement of Artificial Intelligence","author":"Oates T.","year":"2000"},{"key":"bibr27-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1109\/TEVC.2006.890271"},{"key":"bibr28-1059712313511648","volume-title":"Conditioned reflexes: An investigation of the physiological activity of the cerebral cortex","author":"Pavlov I.","year":"1927"},{"key":"bibr29-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2007.11.026"},{"key":"bibr30-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/s11023-008-9095-5"},{"key":"bibr31-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(96)00051-3"},{"key":"bibr32-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1037\/0097-7403.6.3.207"},{"key":"bibr33-1059712313511648","first-page":"202","volume-title":"Proceedings of the Conference of the Association for the Advancement of Artificial Intelligence","author":"Singh S.","year":"1992"},{"key":"bibr34-1059712313511648","first-page":"512","volume-title":"Proceedings of the 20th Conference on Uncertainty in Artificial Intelligence","author":"Singh S.","year":"2004"},{"key":"bibr35-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/BF00115009"},{"key":"bibr36-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1016\/B978-1-55860-141-3.50030-4"},{"key":"bibr37-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1016\/B978-1-55860-377-6.50072-4"},{"key":"bibr38-1059712313511648","volume-title":"Working Notes of the IJCAI-09 Workshop on Grand Challenges for Reasoning from Experiences","author":"Sutton R. S.","year":"2009"},{"key":"bibr39-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-31951-8_2"},{"key":"bibr40-1059712313511648","first-page":"497","volume-title":"Learning and computational neuroscience: Foundations of adaptive networks","author":"Sutton R. S.","year":"1990"},{"key":"bibr41-1059712313511648","volume-title":"Reinforcement learning: An introduction","author":"Sutton R. S.","year":"1998"},{"key":"bibr42-1059712313511648","first-page":"761","volume-title":"Proceedings of the 10th International Conference on Autonomous Agents and Multiagent Systems","author":"Sutton R. S.","year":"2011"},{"key":"bibr43-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1016\/S0004-3702(99)00052-1"},{"key":"bibr44-1059712313511648","author":"Sutton R. S.","year":"2009","journal-title":"Advances in Neural Information Processing Systems 21"},{"key":"bibr45-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1145\/1553374.1553501"},{"key":"bibr46-1059712313511648","first-page":"1377","author":"Sutton R. S.","year":"2005","journal-title":"Advances in Neural Information Processing Systems 17"},{"key":"bibr47-1059712313511648","first-page":"2849","volume-title":"Proceedings of the IEEE\/RSJ International Conference on Intelligent Robots and Systems","author":"Tedrake R.","year":"2005"},{"key":"bibr48-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1002\/rob.20147"},{"key":"bibr49-1059712313511648","volume-title":"Purposive behavior in animals and men","author":"Tolman E. C.","year":"1951"},{"key":"bibr50-1059712313511648","first-page":"842","volume-title":"Proceedings of the IEEE International Conference on Robotics and Automation","author":"Wang C. C.","year":"2003"},{"key":"bibr51-1059712313511648","volume-title":"An Introduction to the Kalman filter","author":"Welch G.","year":"1995"},{"key":"bibr52-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1109\/DevLrn.2012.6400860"},{"key":"bibr53-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1126\/science.7569931"},{"key":"bibr54-1059712313511648","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1000220"}],"container-title":["Adaptive Behavior"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/1059712313511648","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/1059712313511648","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/1059712313511648","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,3]],"date-time":"2025-03-03T00:54:18Z","timestamp":1740963258000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/1059712313511648"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2014,2,7]]},"references-count":54,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2014,4]]}},"alternative-id":["10.1177\/1059712313511648"],"URL":"https:\/\/doi.org\/10.1177\/1059712313511648","relation":{},"ISSN":["1059-7123","1741-2633"],"issn-type":[{"value":"1059-7123","type":"print"},{"value":"1741-2633","type":"electronic"}],"subject":[],"published":{"date-parts":[[2014,2,7]]}}}