A fully analytical recursive stochastic model to the normalized signed regressor LMS algorithm.
Márcio H. CostaJosé C. M. BermudezPublished in: ISSPA (2) (2003)
Keyphrases
- stochastic model
- computational complexity
- times faster
- learning algorithm
- recursive algorithm
- segmentation algorithm
- experimental evaluation
- np hard
- dynamic programming
- least mean square
- detection algorithm
- optimization algorithm
- computational cost
- search space
- cost function
- significant improvement
- k means
- matching algorithm
- similarity measure
- probabilistic model
- probability distribution
- optimization problems
- expectation maximization
- learning process
- steady state
- shape model
- preprocessing
- objective function
- neural network