9 search results - page 1 / 2 » Parallel Coordinate Descent for L1-Regularized Loss Minimiza... |
CORR
2011
Springer
12 years 9 months ago
2011
Springer
We propose Shotgun, a parallel coordinate descent algorithm for minimizing L1regularized losses. Though coordinate descent seems inherently sequential, we prove convergence bounds...
IJCNN
2007
IEEE
13 years 11 months ago
2007
IEEE
—The 0/1 loss is an important cost function for perceptrons. Nevertheless it cannot be easily minimized by most existing perceptron learning algorithms. In this paper, we propose...
JMLR
2012
11 years 7 months ago
2012
We consider the minimization of a smooth loss with trace-norm regularization, which is a natural objective in multi-class and multitask learning. Even though the problem is convex...
JMLR
2008
13 years 5 months ago
2008
Linear support vector machines (SVM) are useful for classifying large-scale sparse data. Problems with sparse features are common in applications such as document classification a...
JMLR
2010
13 years 3 months ago
2010
We extend the well-known BFGS quasi-Newton method and its memory-limited variant LBFGS to the optimization of nonsmooth convex objectives. This is done in a rigorous fashion by ge...