Weakly-correlated synapses promote dimension reduction in deep neural networks.
Jianwen ZhouHaiping HuangPublished in: CoRR (2020)
Keyphrases
- dimension reduction
- neural network
- feature extraction
- high dimensional
- principal component analysis
- high dimensional problems
- associative memory
- linear discriminant analysis
- random projections
- spiking neural networks
- manifold learning
- singular value decomposition
- data mining and machine learning
- pattern recognition
- high dimensionality
- variable selection
- low dimensional
- feature selection
- high dimensional data
- dimensionality reduction
- back propagation
- feature space
- partial least squares
- feed forward
- discriminative information
- preprocessing
- artificial neural networks
- unsupervised learning
- neural network model
- self organizing maps
- synaptic weights
- database
- cluster analysis
- intrinsic dimension
- high dimensional data analysis
- sparse metric learning