Research Article
A Novel Neuron in Kernel Domain
Algorithm 1
The adaptive KNN algorithm (AKNNμ). Complexity for each instance.
Initialization: | learning step | learning kernel width step | primal kernel width | sparsity threshold | ; | While available do | for instances do | % compute distances of dictionary instances to tth instance | ; | % compute kernel vector, output and error for tth instance | | % compute coefficient using (19) | % save for tracking | | % update kernel width σ | | if became negative then | % decrease until becomes non-negative | % update kernel width again. | end if | if () | % add to dictionary and update using (23); | else | % update Coefficients using (24); | end if | | end for | if ( ) | % manipulated step size using (20) | else if () | % exit training process | end if | end while |
|