Description:

SMIDAS is a C++ implementation of the stochastic mirror descent algorithm proposed in

* Shai Shalev-Shwartz and Ambuj Tewari, Stochastic methods for l1 regularized loss minimization. Proceedings of the 26th International Conference on Machine Learning, pages 929-936, 2009.

It can be used for l1-regularized loss minimization for both classification and regression problems.

Currently supported loss functions are the logistic loss, the hinge loss, and the squared loss [L(a,b) = (a-b)2]. SMIDAS is designed to run fast even for high-dimensional large datasets and can exploit the sparsity in the examples.

Changes to previous version:

Fixed major bug in implementation. The components of the iterate where the current example vector is zero were not being updated correctly. Thanks to Jonathan Chang for pointing out the error to us.

BibTeX Entry: Download

Corresponding Paper BibTeX Entry: Download

Supported Operating Systems: Agnostic

Data Formats: Ascii

Tags: L1 Regularization, Large Datasets, Mirror Descent, Sparsity

Archive: download here

Comments

No one has posted any comments yet. Perhaps you'd like to be the first?

Leave a comment

You must be logged in to post comments.