Name
Abstract | ||
|---|---|---|
Classifiers favoring sparse solutions, such as support vector machines, relevance vector machines, LASSO-regression based classifiers, etc., provide competitive methods for classification problems in high dimensions. However, current algorithms for training sparse classifiers typically scale quite unfavorably with respect to the number of training examples. This paper proposes online and multi-pass algorithms for training sparse linear classifiers for high dimensional data. These algorithms have computational complexity and memory requirements that make learning on massive data sets feasible. The central idea that makes this possible is a straightforward quadratic approximation to the likelihood function. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1145/1390681.1390691 | Journal of Machine Learning Research |
Keywords | Field | DocType |
laplace approximation,training example,support vector machine,sparse solution,sparse linear classifiers,high dimension,training sparse,massive data setting,relevance vector machine,sparse linear classifier,lasso,central idea,high dimensional data,massive data,expectation propagation,likelihood function,computational complexity,artificial intelligence,statistics | Data set,Computer science,Random subspace method,Artificial intelligence,Clustering high-dimensional data,Likelihood function,Pattern recognition,Support vector machine,Sparse approximation,Quadratic equation,Algorithm,Machine learning,Computational complexity theory | Journal |
Volume | ISSN | Citations |
9, | 1532-4435 | 30 |
PageRank | References | Authors |
3.76 | 10 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Suhrid Balakrishnan | 1 | 238 | 14.60 |
| David Madigan | 2 | 190 | 22.89 |