Statistical Learning: Stability is Sufficient for Generalization and Necessary and Sufficient for Consistency of Empirical Risk Minimization
dc.date.accessioned | 2004-08-31T18:12:01Z | |
dc.date.accessioned | 2018-11-24T10:09:32Z | |
dc.date.available | 2004-08-31T18:12:01Z | |
dc.date.available | 2018-11-24T10:09:32Z | |
dc.date.issued | 2002-12-01 | en_US |
dc.identifier.uri | http://hdl.handle.net/1721.3/5507 | |
dc.identifier.uri | http://repository.aust.edu.ng/xmlui/handle/1721.3/5507 | |
dc.description | revised July 2003 | en_US |
dc.description.abstract | Solutions of learning problems by Empirical Risk Minimization (ERM) need to be consistent, so that they may be predictive. They also need to be well-posed, so that they can be used robustly. We show that a statistical form of well-posedness, defined in terms of the key property of L-stability, is necessary and sufficient for consistency of ERM. | en_US |
dc.format.extent | 24 p. | en_US |
dc.format.extent | 1854466 bytes | |
dc.format.extent | 400508 bytes | |
dc.language.iso | en_US | |
dc.subject | AI | en_US |
dc.subject | Theory of Learning | en_US |
dc.subject | Great Discoveries | en_US |
dc.subject | Consistency | en_US |
dc.subject | ERM | en_US |
dc.subject | Stability | en_US |
dc.title | Statistical Learning: Stability is Sufficient for Generalization and Necessary and Sufficient for Consistency of Empirical Risk Minimization | en_US |
Files in this item
Files | Size | Format | View |
---|---|---|---|
AIM-2002-024.pdf | 400.5Kb | application/pdf | View/ |
AIM-2002-024.ps | 1.854Mb | application/postscript | View/ |