## Description

**RLS2 MATLAB Toolbox**is a set of scripts that implements

**RLS2**(regularized least squares with two layers) and

**RLS2LIN**(linear regularized least squares with two layers).

RLS2 is an instance of

**multiple kernel learning**algorithm that can be used to simultaneously learn a regularized predictor and the kernel function. It is also an instance of

**kernel machine with two layers**that extends the classic regularized least squares algorithm.

RLS2LIN implements a version of RLS2 specialized to linear kernels on each feature. The algorithm simultaneously performs

**regularization**and linear

**feature selection**, is memory efficient and very well suited for datasets with a large number of features.

The package contains a Graphic User Interface (GUI) to load data, perform training and validation of RLS2 models, and plot results. The features of the toolbox include:

- Data pre-processing.
- Efficient regularization path computation.
- Cross-validation.
- Random splits.
- Hold-out set validation.
- Multi-class classification (one versus all)
- Multi-output regression
- Approximate degrees of freedom computation.
- Plot results and export figures to PDF format.

## References

Francesco Dinuzzo. Kernel machines with two layers and multiple kernel learning. Preprint arXiv:1001.2709.

## Screenshots

## File list

Core:

- rls2.m (Regularized least squares with two layers - training)
- rls2eval.m (Regularized least squares with two layers - test)
- rls2lin.m (Regularized least squares linear with two layers - training)
- rls2lineval.m (Regularized least squares linear with two layers - test)
- gpl.txt (GNU license)
- readme.txt

- rls2tools.m (Graphic User Interface for RLS2 and RLS2LIN - .m file)
- rls2tools.fig (Graphic User Interface for RLS2 and RLS2LIN - .fig file)

- kernels/default.m (Example of function computing a set of basis kernel matrices.)
- kernels/rbfall.m (RBF Kernels on all the features)
- kernels/polyall.m (Polynomial kernels on all the features)
- kernels/rbfsingle.m (RBF Kernels on each feature separately)
- kernels/polysingle.m (Polynomial kernels on each feature separately)

- data/heart.mat (Example classification dataset)
- data/housing.mat (Example regression dataset)
- data/iris.mat (Example multi-classification dataset)
- data/prostate.mat (Example regression dataset with pre-defined validation set)
- data/binarystrings.mat (Regression dataset with inputs in logical format)

- doc/tutorial.pdf (Tutorial)

## Installation

To install the toolbox, simply unpack RLS2.zip into some folder and add that folder to the MATLAB path, by selecting "Set Path..." from the File menu. To enable the

**Generate PDF plots**feature of the GUI, you need to download the script save2pdf.m and save it into the Matlab path.## Disclaimer

Copyright © 2010 Francesco Dinuzzo

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.

## Download

Please read the license before downloading the toolbox.