No description, website, or topics provided.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
python
README.md
a9a
comp_logistic_diag.m
comp_logistic_grad.m
demo_comparison.m
fig_result.m
lbfgs.m
libsvmread.mexmaci64
logistic_loss.m
newton_solver.m
subsampled_newton.m

README.md

Subsampled-Newton

Codes for subsampled Newton methods for solving ridge logistic regression.

About

Consider minimizing a sum of convex functions. Subsampled-Newton methods subsample the functions to calcuate the approximated Hessian. Using non-uniform sampling schemes, we can show the sampling size can be independent of number of functions. In typical ERM problems where n >> d, sub-sampled Newton methods can speed up a lot.

Usage

  • Matlab users: See subsampled_newton.m for main functions. And see demo_comparison.m for usage.

  • Python users: See files in the folder python.

Reference

Peng Xu, Jiyan Yang, Farbod Roosta-Khorasani, Christopher Ré, and Michael W. Mahoney, Sub-sampled Newton Methods with Non-uniform Sampling, NIPS 2016.