P95-1031 Inside-Outside algorithm , the gradient descent search discovers the " nearest " local
D09-1152 training set . We then perform a gradient descent search to find optimal settings for
W07-1605 different models and suggest a gradient descent search for parameter esti - mation .
P94-1004 path can be used to constrain a gradient descent search , called the forwardbackward
W97-0302 We . de - signed a variant on a gradient descent search algorithm to find the optimal
J13-3007 hypernyms for backpropagation are gradient descent search procedure and training algorithm
W07-1605 are scaled logarithmically . A gradient descent search on the basis of the ZM model
C94-1027 gation algorithm which performs a gradient descent search on the error surface , The weight
hide detail