A class of fast quaternion valued variable stepsize stochastic gradient learning algorithms for vector sensor processes.
Mingxuan WangClive Cheong TookDanilo P. MandicPublished in: IJCNN (2011)
Keyphrases
- stochastic gradient
- step size
- learning algorithm
- stochastic gradient descent
- convergence rate
- convergence speed
- cost function
- learning rate
- machine learning algorithms
- training data
- learning problems
- feature vectors
- active learning
- nearest neighbor classifier
- wavelet coefficients
- machine learning
- similarity measure
- learning tasks
- loss function
- collaborative filtering
- supervised learning
- data points
- dynamic programming
- multiresolution