To tune or not to tune: Recommending when to adjust SVM hyper-parameters via meta-learning.
Rafael Gomes MantovaniAndré Luis Debiaso RossiJoaquin VanschorenBernd BischlAndré C. P. L. F. de CarvalhoPublished in: IJCNN (2015)
Keyphrases
- meta learning
- model selection
- hyperparameters
- grid search
- support vector
- feature selection
- cross validation
- inductive learning
- learning tasks
- machine learning
- svm classifier
- support vector machine svm
- decision trees
- sample size
- support vector machine
- closed form
- bayesian framework
- data mining
- training data
- kernel methods
- feature space
- kernel function
- bayesian networks
- incremental learning
- metamodel
- base classifiers
- bayesian inference
- feature extraction
- feature set