A Fast Neural Network Learning Algorithm with Approximate Singular Value Decomposition.
Norbert JankowskiRafal LinowieckiPublished in: Int. J. Appl. Math. Comput. Sci. (2019)
Keyphrases
- singular value decomposition
- neural network
- learning algorithm
- back propagation
- singular values
- least squares
- low rank
- dimensionality reduction
- latent semantic indexing
- training data
- dimension reduction
- error minimization
- machine learning
- principal component analysis
- latent semantic analysis
- active learning
- correlation matrix
- low rank approximation
- reinforcement learning
- pattern recognition
- supervised learning
- policy evaluation
- low rank matrix
- singular vectors
- structure from motion
- computer vision
- data sets
- data matrix
- moving objects
- knn
- high dimensional