NIPS
2004
13 years 5 months ago
2004
We describe an algorithm for support vector machines (SVM) that can be parallelized efficiently and scales to very large problems with hundreds of thousands of training vectors. I...