# bmrm v4.1

0

0th

Percentile

## Bundle Methods for Regularized Risk Minimization Package

Bundle methods for minimization of convex and non-convex risk under L1 or L2 regularization. Implements the algorithm proposed by Teo et al. (JMLR 2010) as well as the extension proposed by Do and Artieres (JMLR 2012). The package comes with lot of loss functions for machine learning which make it powerful for big data analysis. The applications includes: structured prediction, linear SVM, multi-class SVM, f-beta optimization, ROC optimization, ordinal regression, quantile regression, epsilon insensitive regression, least mean square, logistic regression, least absolute deviation regression (see package examples), etc... all with L1 and L2 regularization.

## Functions in bmrm

 Name Description is.convex Return or set is.convex attribute balanced.cv.fold Split a dataset for Cross Validation taking into account class balance bhattacharyya.coefficient Compute Bhattacharyya coefficient needed for Hellinger distance linearRegressionLoss Loss functions to perform a regression hellinger.dist Compute Hellinger distance hclust_fca Find first common ancestor of 2 nodes in an hclust object balanced.loss.weights Compute loss.weights so that total losses of each class is balanced ordinalRegressionLoss The loss function for ordinal regression predict.mmc Predict class of new instances according to a mmc model rank.linear.weights Rank linear weight of a linear model iterative.hclust Perform multiple hierachical clustering on random subsets of a dataset roc.stat Compute statistics for ROC curve plotting mmcLoss Loss function for max-margin clustering preferenceLoss The loss function for Preference loss lpSVM Linearly Programmed SVM print.roc.stat Generic method overlad to print object of class roc.stat multivariateHingeLoss The loss function for multivariate hinge loss softmaxLoss softmax Loss Function binaryClassificationLoss Loss functions for binary classification wolfe.linesearch Wolfe Line Search lvalue Return or set lvalue attribute mmc Convenient wrapper function to solve max-margin clustering problem on a dataset rowmean Columun means of a matrix based on a grouping variable softMarginVectorLoss Soft Margin Vector Loss function for multiclass SVM costMatrix Compute or check the structure of a cost matrix gradient Return or set gradient attribute nrbm Convex and non-convex risk minimization with L2 regularization and limited memory ontologyLoss Ontology Loss Function No Results!