The landscape of empirical risk for nonconvex losses (Q1991675): Difference between revisions
From MaRDI portal
Created a new Item |
Added link to MaRDI item. |
||
links / mardi / name | links / mardi / name | ||
Revision as of 17:09, 1 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | The landscape of empirical risk for nonconvex losses |
scientific article |
Statements
The landscape of empirical risk for nonconvex losses (English)
0 references
30 October 2018
0 references
In study of the complexity of M-estimators, the landscape of the empirical risk, namely its stationary points and their properties are examined. Uniform convergence of the gradient and Hessian of the empirical risk to their population counterparts is established, as soon as the number of samples becomes larger than the number of unknown parameters (modulo logarithmic factors). A characterization of the empirical risk landscape under a nearly information-theoretically minimal condition is provided. The result is as follows, if the number of samples exceeds the sparsity of the parameter vector (modulo logarithmic factor), then a suitable uniform convergence result holds. The paper is organized as follows: good definitions, nice theorems, attractive examples, such as nonconvex binary classification and robust regression in very high dimension.
0 references
nonconvex optimization
0 references
empirical risk minimization
0 references
landscape
0 references
uniform convergence
0 references