Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/9055
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | VAN DEN BROECK, Christian | - |
dc.date.accessioned | 2008-12-19T10:42:41Z | - |
dc.date.available | 2008-12-19T10:42:41Z | - |
dc.date.issued | 1998 | - |
dc.identifier.citation | Wong, KYM. & King, I. & Yeung, DY. (Ed.) THEORETICAL ASPECTS OF NEURAL COMPUTATION - A MULTIDISCIPLINARY PERSPECTIVE. p. 249-255. | - |
dc.identifier.uri | http://hdl.handle.net/1942/9055 | - |
dc.description.abstract | We study both on-line and off-line learning in the following unsupervised learning scheme: p patterns are sampled independently from a distribution on the N-sphere with a single symmetry breaking orientation. Exact results are obtained in the limit p --> infinity and N --> infinity with finite ratio p/N. One finds that for smooth pattern distributions, the asymptotic behavior of the optimal off-line and on-line learning are identical, and saturate the Cramer-Rao inequality from statistics. For discontinuous pattern distributions on the other hand, the optimal online algorithm needs (at least) twice as many examples asymptotically to reach the optimal off-line performance. | - |
dc.language.iso | en | - |
dc.publisher | SPRINGER-VERLAG SINGAPORE PTE LTD | - |
dc.title | Unsupervised learning by examples: On-line versus off-line | - |
dc.type | Proceedings Paper | - |
local.bibliographicCitation.authors | Wong, KYM. | - |
local.bibliographicCitation.authors | King, I. | - |
local.bibliographicCitation.authors | Yeung, DY. | - |
local.bibliographicCitation.conferencename | Hong Kong International Workshop on Theoretical Aspects of Neural Computation - A Multi-Disciplinary Perspective (TANC-97) | - |
local.bibliographicCitation.conferenceplace | HONG KONG, MAY 26-28, 1997 | - |
dc.identifier.epage | 255 | - |
dc.identifier.spage | 249 | - |
local.format.pages | 7 | - |
dc.description.notes | Limburgs Univ Ctr, B-3590 Diepenbeek, Belgium. | - |
local.type.refereed | Refereed | - |
local.type.specified | Proceedings Paper | - |
dc.bibliographicCitation.oldjcat | C1 | - |
dc.identifier.isi | 000078895600023 | - |
local.bibliographicCitation.btitle | THEORETICAL ASPECTS OF NEURAL COMPUTATION - A MULTIDISCIPLINARY PERSPECTIVE | - |
item.fulltext | No Fulltext | - |
item.accessRights | Closed Access | - |
item.contributor | VAN DEN BROECK, Christian | - |
item.validation | ecoom 2000 | - |
item.fullcitation | VAN DEN BROECK, Christian (1998) Unsupervised learning by examples: On-line versus off-line. In: Wong, KYM. & King, I. & Yeung, DY. (Ed.) THEORETICAL ASPECTS OF NEURAL COMPUTATION - A MULTIDISCIPLINARY PERSPECTIVE. p. 249-255.. | - |
Appears in Collections: | Research publications |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.