A new method for building single feedforward neural network models for multivariate static regression problems: a combined weight initialization and constructive algorithm.
Ghabriel A. Gomes de SáCristiano Hora FontesMarcelo EmbiruçuPublished in: Evol. Intell. (2024)
Keyphrases
- regression problems
- k means
- feed forward
- linear regression
- support vector machine svm
- genetic programming
- cross validation
- convergence rate
- energy function
- neural nets
- artificial neural networks
- neural network model
- learning algorithm
- bayesian framework
- prior information
- prior knowledge
- support vector machine
- input data
- probabilistic model
- hyperparameters
- closed form
- machine learning
- multi class
- model selection
- feature extraction
- feature selection
- bayesian inference
- random forests