Learning Machines: Foundations of Trainable Pattern-classifying SystemsMcGraw-Hill, 1965 - 137 sivua |
Kirjan sisältä
Tulokset 1 - 3 kokonaismäärästä 40
Sivu 70
... W responds incorrectly to an augmented pattern vector Y. The weight vec- tor is then changed to a new weight vector W ' , as follows : W ' W ' = W + CY = W - CY if Y belonged to category 1 if Y belonged to category 2 ( 4.5 ) where c is ...
... W responds incorrectly to an augmented pattern vector Y. The weight vec- tor is then changed to a new weight vector W ' , as follows : W ' W ' = W + CY = W - CY if Y belonged to category 1 if Y belonged to category 2 ( 4.5 ) where c is ...
Sivu 85
... W in W and each Y in y ' Here W is an open convex region bounded by hyperplanes ( the pattern hyperplanes ) all of which pass through the origin . Such a region is called a convex polyhedral cone with vertex at the origin . If a ...
... W in W and each Y in y ' Here W is an open convex region bounded by hyperplanes ( the pattern hyperplanes ) all of which pass through the origin . Such a region is called a convex polyhedral cone with vertex at the origin . If a ...
Sivu 92
... W≤ Ŵx - W ( 5.38 ) for all W in w . We therefore say that Ŵ + 1 is pointwise closer than Ŵ to W. As a first step in proving the theorem , we shall show that the sequence S✩ converges to a point P. For any fixed W in W let lim | Ŵ — W ] = ...
... W≤ Ŵx - W ( 5.38 ) for all W in w . We therefore say that Ŵ + 1 is pointwise closer than Ŵ to W. As a first step in proving the theorem , we shall show that the sequence S✩ converges to a point P. For any fixed W in W let lim | Ŵ — W ] = ...
Sisältö
TRAINABLE PATTERN CLASSIFIERS | 1 |
PARAMETRIC TRAINING METHODS | 43 |
SOME NONPARAMETRIC TRAINING METHODS | 65 |
Tekijänoikeudet | |
2 muita osia ei näytetty
Muita painoksia - Näytä kaikki
Yleiset termit ja lausekkeet
adjusted apply assume bank belonging to category called changes Chapter cluster committee components consider consists contains correction corresponding decision surfaces define denote density depends derivation described Development discriminant functions discussed distance distribution element equal error-correction estimates example exists expression FIGURE fixed given implemented important initial layered machine linear dichotomies linear discriminant functions linear machine linearly separable measurements negative networks normal Note optimum origin parameters partition pattern classifier pattern hyperplane pattern space pattern vector piecewise linear plane points positive presented probability problem proof properties proved PWL machine quadric reduced regions respect response rule sample mean selection separable shown side space Stanford step subsidiary discriminant Suppose theorem theory threshold training methods training procedure training sequence training subsets transformation values weight vectors zero
Viitteet tähän teokseen
A Probabilistic Theory of Pattern Recognition Luc Devroye,László Györfi,Gabor Lugosi Rajoitettu esikatselu - 1997 |