Training Echo State Networks with Regularization Through Dimensionality Reduction.
Sigurd LøkseFilippo Maria BianchiRobert JenssenPublished in: Cogn. Comput. (2017)
Keyphrases
- echo state networks
- dimensionality reduction
- recurrent neural networks
- high dimensional data
- high dimensional
- feature extraction
- early stopping
- principal component analysis
- pattern recognition and machine learning
- data sets
- principal components
- structure preserving
- low dimensional
- pattern recognition
- linear dimensionality reduction
- nonlinear dimensionality reduction
- kernel pca
- regularization parameter
- image processing
- data representation
- high dimensionality
- support vector machine
- training set
- input space
- manifold learning
- cross validation
- training samples
- online learning
- pairwise