Please use this identifier to cite or link to this item:
https://doi.org/10.1006/jcss.1996.0033
DC Field | Value | |
---|---|---|
dc.title | Fat-shattering and the learnability of real-valued functions | |
dc.contributor.author | Bartlett, P.L. | |
dc.contributor.author | Long, P.M. | |
dc.contributor.author | Williamson, R.C. | |
dc.date.accessioned | 2014-10-27T06:02:34Z | |
dc.date.available | 2014-10-27T06:02:34Z | |
dc.date.issued | 1996-06 | |
dc.identifier.citation | Bartlett, P.L., Long, P.M., Williamson, R.C. (1996-06). Fat-shattering and the learnability of real-valued functions. Journal of Computer and System Sciences 52 (3) : 434-452. ScholarBank@NUS Repository. https://doi.org/10.1006/jcss.1996.0033 | |
dc.identifier.issn | 00220000 | |
dc.identifier.uri | http://scholarbank.nus.edu.sg/handle/10635/99288 | |
dc.description.abstract | We consider the problem of learning real-valued functions from random examples when the function values are corrupted with noise. With mild conditions on independent observation noise, we provide characterizations of the learnability of a real-valued function class in terms of a generalization of the Vapnik-Chervonenkis dimension, the fat-shattering function, introduced by Kearns and Schapire. We show that, given some restrictions on the noise, a function class is learnable in our model if an only if its fat-shattering function is finite. With different (also quite mild) restrictions, satisfied for example by guassion noise, we show that a function class is learnable from polynomially many examples if and only if its fat-shattering function grows polynomially. We prove analogous results in an agnostic setting, where there is no assumption of an underlying function class. © 1996 Academic Press, Inc. | |
dc.description.uri | http://libproxy1.nus.edu.sg/login?url=http://dx.doi.org/10.1006/jcss.1996.0033 | |
dc.source | Scopus | |
dc.type | Article | |
dc.contributor.department | INFORMATION SYSTEMS & COMPUTER SCIENCE | |
dc.description.doi | 10.1006/jcss.1996.0033 | |
dc.description.sourcetitle | Journal of Computer and System Sciences | |
dc.description.volume | 52 | |
dc.description.issue | 3 | |
dc.description.page | 434-452 | |
dc.description.coden | JCSSB | |
dc.identifier.isiut | A1996UW48900005 | |
Appears in Collections: | Staff Publications |
Show simple item record
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.