A sequential dual method for large scale multi-class linear svms
S. Sathiya Keerthi, Srikrishnan Sundararajan, Kai‐Wei Chang et al. · ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD) · 2008
Efficient training of direct multi-class formulations of linear Support Vector Machines is very useful in applications such as text classification with a huge number examples as well as features. This paper presents a fast dual method for this training. The main idea is to sequentially traverse through the training set and optimize the dual variables associated with one example at a time. The speed of training is enhanced further by shrinking and cooling heuristics. Experiments indicate that our method is much faster than state of the art solvers such as bundle, cutting plane and exponentiated gradient methods.
Cited by 149 works. doi:10.1145/1401890.1401942
Authors
References
- Gradient-based learning applied to document recognition
- A training algorithm for optimal margin classifiers
- Making Large-Scale SVM Learning Practical
- Practical Methods of Optimization
- Advances in kernel methods: support vector learning
- Practical methods of optimization
- A comparison of methods for multiclass support vector machines
- Practical Methods of Optimization.
- A comparison of event models for naive bayes text classification
- RCV1: A New Benchmark Collection for Text Categorization Research