Non-Euclidean Principal Component Analysis and Oja's Learning Rule - Theoretical Aspects.
Michael BiehlMarika KästnerMandy LangeThomas VillmannPublished in: WSOM (2012)
Keyphrases
- learning rules
- principal component analysis
- neural network
- biologically inspired
- principal components
- multi dimensional scaling
- independent component analysis
- artificial neural networks
- dimensionality reduction
- euclidean space
- low dimensional
- face recognition
- learning scheme
- covariance matrix
- back propagation
- linear discriminant analysis
- feature space
- learning mechanism
- learning algorithm
- spiking neural networks
- euclidean distance
- face images
- feature extraction
- genetic algorithm
- receptive fields
- biological systems
- fuzzy logic
- decision making