Sharper lower bounds on the performance of the empirical risk minimization algorithm
dc.contributor.author | Lecue, G | |
dc.contributor.author | Mendelson, Shahar | |
dc.date.accessioned | 2015-12-10T22:31:51Z | |
dc.date.issued | 2010 | |
dc.date.updated | 2016-02-24T08:27:32Z | |
dc.description.abstract | We present an argument based on the multidimensional and the uniform central limit theorems, proving that, under some geometrical assumptions between the target function T and the learning class F, the excess risk of the empirical risk minimization algorithm is lower bounded by Esup q∈Q Gq/δ,/n where (Gq)q∈Q is a canonical Gaussian process associated with Q (a well chosen subset of F) and δ is a parameter governing the oscillations of the empirical excess risk function over a small ball in F. | |
dc.identifier.issn | 1350-7265 | |
dc.identifier.uri | http://hdl.handle.net/1885/55533 | |
dc.publisher | Chapman & Hall | |
dc.source | Bernoulli | |
dc.subject | Keywords: Empirical risk minimization; Learning theory; Lower bound; Multidimensional central limit theorem; Uniform central limit theorem | |
dc.title | Sharper lower bounds on the performance of the empirical risk minimization algorithm | |
dc.type | Journal article | |
local.bibliographicCitation.issue | 3 | |
local.bibliographicCitation.lastpage | 613 | |
local.bibliographicCitation.startpage | 605 | |
local.contributor.affiliation | Lecue, G, CNRS | |
local.contributor.affiliation | Mendelson, Shahar, College of Physical and Mathematical Sciences, ANU | |
local.contributor.authoremail | u4011413@anu.edu.au | |
local.contributor.authoruid | Mendelson, Shahar, u4011413 | |
local.description.embargo | 2037-12-31 | |
local.description.notes | Imported from ARIES | |
local.identifier.absfor | 080201 - Analysis of Algorithms and Complexity | |
local.identifier.ariespublication | f2965xPUB332 | |
local.identifier.citationvolume | 16 | |
local.identifier.doi | 10.3150/09-BEJ225 | |
local.identifier.scopusID | 2-s2.0-77957597566 | |
local.identifier.uidSubmittedBy | f2965 | |
local.type.status | Published Version |
Downloads
Original bundle
1 - 1 of 1
Loading...
- Name:
- 01_Lecue_Sharper_lower_bounds_on_the_2010.pdf
- Size:
- 112.87 KB
- Format:
- Adobe Portable Document Format