Neural-Gas for Function Approximation: A Heuristic for Minimizing the Local Estimation Error.
M. WinterGiorgio MettaGiulio SandiniPublished in: IJCNN (4) (2000)
Keyphrases
- function approximation
- estimation error
- neural gas
- self organizing maps
- reinforcement learning
- vector quantization
- learning tasks
- radial basis function
- color quantization
- vector quantizer
- error rate
- standard deviation
- cluster analysis
- approximation error
- covariance matrix
- dynamic programming
- optimal solution
- neural network
- image segmentation
- step size
- image processing
- competitive learning
- training data
- unsupervised learning
- image compression
- color images
- computational complexity
- genetic algorithm