A novel LS-SVMs hyper-parameter selection based on particle swarm optimization.
X. C. GuoJ. H. YangC. G. WuC. Y. WangY. C. LiangPublished in: Neurocomputing (2008)
Keyphrases
- particle swarm optimization
- hyperparameters
- ls svm
- cross validation
- pso algorithm
- bayesian inference
- closed form
- genetic algorithm
- random sampling
- bayesian framework
- differential evolution
- model selection
- support vector
- particle swarm optimization pso
- em algorithm
- particle swarm
- image segmentation
- convergence speed
- evolutionary algorithm
- high dimensional
- multi objective
- sample size
- machine learning
- higher order
- prior knowledge
- prior information
- incremental learning
- parameter settings
- multi class
- variable selection
- decision trees
- feature selection
- least squares support vector machine