Response surface methodology to tune artificial neural network hyper-parameters.
Sinem Bozkurt KeserYeliz Buruk SahinPublished in: Expert Syst. J. Knowl. Eng. (2021)
Keyphrases
- hyperparameters
- response surface methodology
- artificial neural networks
- anfis model
- model selection
- support vector regression
- support vector
- neural network
- prediction model
- cross validation
- bayesian inference
- closed form
- random sampling
- bayesian framework
- prior information
- particle swarm optimisation
- noise level
- em algorithm
- sample size
- regression model
- maximum likelihood
- incremental learning
- back propagation
- maximum a posteriori
- genetic algorithm ga
- neuro fuzzy
- genetic algorithm
- incomplete data
- parameter settings
- input variables
- parameter optimization
- bp neural network
- linear regression model
- experimental data
- prior knowledge
- variable selection
- fuzzy inference system
- data sets
- probabilistic model
- training data
- missing values
- recurrent neural networks
- noisy images
- parameter space
- expectation maximization
- level set
- support vector machine
- machine learning
- active learning