DALhttp://mloss.orgUpdates and additions to DALenTue, 18 Feb 2014 19:07:06 -0000DAL 1.1<html><ul> <li><p>DAL is an efficient and flexibible MATLAB toolbox for solving the following optimization problem: </p> <pre><code>minimize f(Ax) + lambda*c(x) </code></pre><p>where A (m x n) is a design matrix, f is a loss function, and c is a measure of sparsity. </p> </li> <li><p>DAL can handle your favorite (convex, smooth) loss functions (squared loss, logistic loss, etc). </p> </li> <li><p>DAL can handle A (and its transpose) provided as function handles. </p> </li> <li><p>DAL can handle several "sparsity" measures in an unified way. Currently L1, grouped L1, and trace norm (testing, requires PROPACK) measures are supported. </p> </li> <li><p>DAL is efficient when m&lt;&lt;n (m: #samples, n: #unknowns) or the matrix A is poorly conditioned. </p> </li> <li><p>DAL is fast when the solution is sparse but the matrix A can be dense. </p> </li> <li><p>DAL is written in MATLAB. </p> </li> </ul></html>Ryota TomiokaTue, 18 Feb 2014 19:07:06 -0000 normgroup lassolassosparse learningl1 regularizationlogistic regression