A new Meta Machine Learning (MML) method based on combining non-significant different neural networks.
Andrés Yáñez EscolanoJoaquín Pizarro JunqueraElisa Guerrero VázquezPedro Galindo RiañoPublished in: ESANN (2003)
Keyphrases
- neural network
- machine learning
- detection method
- experimental evaluation
- high accuracy
- cost function
- significant improvement
- machine learning methods
- high precision
- synthetic data
- support vector machine
- computer vision
- bp neural network
- unsupervised learning
- model selection
- text classification
- fuzzy logic
- computational cost
- mixture model
- detection algorithm
- learning algorithm
- reinforcement learning
- energy function
- segmentation method
- classification method
- dynamic programming
- pairwise
- artificial neural networks