Paper Info
Title
Tightening the Sample Complexity of Empirical Risk Minimization via Preconditioned Stability.
Abstract
We tighten the sample complexity of empirical risk minimization (ERM) associated with a class of generalized linear models that include linear and logistic regression. In particular, we conclude that ERM attains the optimal sample complexity for linear regression. Our analysis relies on a new notion of stability, called preconditioned stability, which may be of independent interest.
Year
Venue
Field
2016
arXiv: Learning
Econometrics,Mathematical optimization,Empirical risk minimization,Generalized linear model,Sample complexity,Logistic regression,Mathematics,Linear regression
DocType
Volume
Citations 
Journal
abs/1601.04011
0
PageRank 
References 
Authors
0.34
0
2
Name
Order
Citations
PageRank
Alon Gonen11049.76
Shai Shalev-Shwartz23681276.32