Can I read Improved Parallel Algorithms For Sequential Minimal Optimization of Classification Problems on EtoBox?
Improved Parallel Algorithms For Sequential Minimal Optimization of Classification Problems by dridabs is a document available to read on EtoBox.
What is Improved Parallel Algorithms For Sequential Minimal Optimization of Classification Problems about?
The paper presents two novel parallel algorithms to improve the convergence of Sequential Minimal Optimization (SMO) for classification problems. The first algorithm, IPSMO-1, enhances convergence by selecting multiple violating pairs in each iteration, while the second, IPSMO-2, further optimizes the selection process to reduce the number of iterations. Experimental results demonstrate that both algorithms effectively execute in parallel and guarantee convergence, outperforming previous methods.
- Author
- dridabs
- Language
- EN