About: A Matlab implementation of Sparse PCA using the inverse power method for nonlinear eigenproblems.

Initial Announcement on mloss.org.

About: A stochastic variant of the mirror descent algorithm employing Langford and Zhang's truncated gradient idea to minimize L1 regularized loss minimization problems for classification and regression.

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.

About: RLS2 is an instance of multiple kernel learning algorithm to simultaneously learn a regularized predictor and the kernel function. RLS2LIN is a version of RLS2 specialized to linear kernels on each feature. The package contains a set of scripts that implements RLS2 and RLS2LIN, together with a Graphic User Interface to load data, perform training, validation, and plot results.

• New kernel functions (rbfall, rbfsingle, polyall, polysingle)
• Improved interface for pre-processing operations
• The interface now allows to disable bias
• Fixed bugs in parameter passing (thanks to Andrea Schirru)