{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T04:47:02Z","timestamp":1750308422806,"version":"3.41.0"},"reference-count":28,"publisher":"Association for Computing Machinery (ACM)","issue":"6","license":[{"start":{"date-parts":[[2019,11,23]],"date-time":"2019-11-23T00:00:00Z","timestamp":1574467200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["J. ACM"],"published-print":{"date-parts":[[2019,12,31]]},"abstract":"<jats:p>\n            We study learning problems involving arbitrary classes of functions\n            <jats:italic>F<\/jats:italic>\n            , underlying measures \u03bc, and targets\n            <jats:italic>Y<\/jats:italic>\n            . Because\n            <jats:italic>proper<\/jats:italic>\n            learning procedures, i.e., procedures that are only allowed to select functions in\n            <jats:italic>F<\/jats:italic>\n            , tend to perform poorly unless the problem satisfies some additional structural property (e.g., that\n            <jats:italic>F<\/jats:italic>\n            is convex), we consider\n            <jats:italic>unrestricted learning procedures<\/jats:italic>\n            that are free to choose functions outside the given class.\n          <\/jats:p>\n          <jats:p>\n            We present a new unrestricted procedure whose sample complexity is almost the best that one can hope for and holds for (almost) any problem, including heavy-tailed situations. Moreover, the sample complexity coincides with what one could expect if\n            <jats:italic>F<\/jats:italic>\n            were convex, even when\n            <jats:italic>F<\/jats:italic>\n            is not. And if\n            <jats:italic>F<\/jats:italic>\n            is convex, then the unrestricted procedure turns out to be proper.\n          <\/jats:p>","DOI":"10.1145\/3361699","type":"journal-article","created":{"date-parts":[[2019,11,25]],"date-time":"2019-11-25T13:19:46Z","timestamp":1574687986000},"page":"1-42","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":6,"title":["An Unrestricted Learning Procedure"],"prefix":"10.1145","volume":"66","author":[{"given":"Shahar","family":"Mendelson","sequence":"first","affiliation":[{"name":"Mathematical Sciences Institute, The Australian National University, Australia and LPSM, Sorbonne University, France"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2019,11,23]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"crossref","unstructured":"M. Anthony and P. L. Bartlett. 1999. Neural Network Learning: Theoretical Foundations. Cambridge University Press Cambridge.","DOI":"10.1017\/CBO9780511624216"},{"key":"e_1_2_1_2_1","unstructured":"J.-Y. Audibert. 2007. Proof of the optimality of the empirical star algorithm. (unpublished)."},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1214\/08-AOS623"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1214\/009053605000000282"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1093\/acprof:oso\/9780199535255.001.0001"},{"key":"e_1_2_1_6_1","doi-asserted-by":"crossref","unstructured":"L. Devroye L. Gy\u00f6rfi and G. Lugosi. 1996. A Probabilistic Theory of Pattern Recognition volume 31 of Applications of Mathematics (New York). Springer-Verlag New York.","DOI":"10.1007\/978-1-4612-0711-5"},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1006\/inco.1994.1097"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1007\/BF02761698"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1214\/07-AOS546"},{"key":"e_1_2_1_10_1","series-title":"Lecture Notes in Mathematics","volume-title":"Oracle Inequalities in Empirical Risk Minimization and Sparse Recovery Problems","author":"Koltchinskii V.","unstructured":"V. Koltchinskii. Oracle Inequalities in Empirical Risk Minimization and Sparse Recovery Problems, volume 2033 of Lecture Notes in Mathematics."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.4064\/sm223-3-5"},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00440-008-0180-8"},{"key":"e_1_2_1_13_1","unstructured":"G. Lecu\u00e9 and S. Mendelson. 2013. Learning subgaussian classes: Upper and minimax bounds. https:\/\/arxiv.org\/abs\/1305.4825 2013."},{"key":"e_1_2_1_14_1","doi-asserted-by":"crossref","unstructured":"M. Ledoux and M. Talagrand. 1991. Probability in Banach Space. Springer-Verlag New York.","DOI":"10.1007\/978-3-642-20212-4"},{"key":"e_1_2_1_15_1","first-page":"1260","article-title":"Learning with square loss: Localization through offset rademacher complexity","volume":"40","author":"Liang T.","year":"2015","unstructured":"T. Liang, A. Rakhlin, and K. Sridharan. 2015. Learning with square loss: Localization through offset rademacher complexity. Proc. Mach. Learn. Res. 40 (2015), 1260--1285.","journal-title":"Proc. Mach. Learn. Res."},{"key":"e_1_2_1_16_1","unstructured":"G. Lugosi and S. Mendelson. Risk minimization by median-of-means tournaments. Journal of the European Mathematical Society."},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jco.2007.09.001"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/2699439"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.spa.2016.04.019"},{"key":"e_1_2_1_20_1","unstructured":"S. Mendelson. 2017. Extending the scope of the small-ball method (unpublished)."},{"key":"e_1_2_1_21_1","first-page":"1","article-title":"Learning without concentration for general loss functions","volume":"171","author":"Mendelson S.","year":"2017","unstructured":"S. Mendelson. 2017. Learning without concentration for general loss functions. Probab. Theory Relat. Fields 171, 1--2 (2017), 459--502.","journal-title":"Probab. Theory Relat. Fields"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1214\/16-AOS1510"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00440-016-0720-6"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.3150\/14-BEJ679"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1214\/aop\/1176988847"},{"key":"e_1_2_1_26_1","series-title":"Springer Series in Statistics","volume-title":"Introduction to Nonparametric Estimation","author":"Tsybakov A. B.","unstructured":"A. B. Tsybakov. Introduction to Nonparametric Estimation. Springer Series in Statistics."},{"key":"e_1_2_1_27_1","doi-asserted-by":"crossref","unstructured":"A. W. van der Vaart and J. A. Wellner. 1996. Weak Convergence and Empirical Processes. Springer-Verlag New York.","DOI":"10.1007\/978-1-4757-2545-2"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.tcs.2017.12.029"}],"container-title":["Journal of the ACM"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3361699","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3361699","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T17:49:27Z","timestamp":1750268967000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3361699"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,11,23]]},"references-count":28,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2019,12,31]]}},"alternative-id":["10.1145\/3361699"],"URL":"https:\/\/doi.org\/10.1145\/3361699","relation":{},"ISSN":["0004-5411","1557-735X"],"issn-type":[{"type":"print","value":"0004-5411"},{"type":"electronic","value":"1557-735X"}],"subject":[],"published":{"date-parts":[[2019,11,23]]},"assertion":[{"value":"2018-03-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-09-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-11-23","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}