Learning Machines: Foundations of Trainable Pattern-classifying SystemsMcGraw-Hill, 1965 - 137 sivua |
Kirjan sisältä
Tulokset 1 - 3 kokonaismäärästä 13
Sivu 75
... product of a weight vector with an augmented pattern vector ; that is , gi ( X ) = W ( i ) . Y for i = 1 , R ( 4.7 ) Simple extensions of the training procedures already discussed can be used to train a general linear machine . Suppose ...
... product of a weight vector with an augmented pattern vector ; that is , gi ( X ) = W ( i ) . Y for i = 1 , R ( 4.7 ) Simple extensions of the training procedures already discussed can be used to train a general linear machine . Suppose ...
Sivu 85
... solution region of weight vectors satisfying inequality ( 5.6 ) . That is , W.Y > 0 for all W in W and each Y in y ... solution region W exists , then clearly each W in W can be scaled such that its dot product with each of the members ...
... solution region of weight vectors satisfying inequality ( 5.6 ) . That is , W.Y > 0 for all W in W and each Y in y ... solution region W exists , then clearly each W in W can be scaled such that its dot product with each of the members ...
Sivu 88
... weight vectors ; Y belongs to one of the train- ing subsets , say Yi . or Then , either ( a ) W , ( k ) . Yk > W1 ... solution weight . vectors . That is , for some set of indices k1 , k2 , KR , the set of vectors " { W , ( k , ) = W2 ...
... weight vectors ; Y belongs to one of the train- ing subsets , say Yi . or Then , either ( a ) W , ( k ) . Yk > W1 ... solution weight . vectors . That is , for some set of indices k1 , k2 , KR , the set of vectors " { W , ( k , ) = W2 ...
Sisältö
TRAINABLE PATTERN CLASSIFIERS | 1 |
PARAMETRIC TRAINING METHODS | 43 |
SOME NONPARAMETRIC TRAINING METHODS | 65 |
Tekijänoikeudet | |
3 muita osia ei näytetty
Muita painoksia - Näytä kaikki
Yleiset termit ja lausekkeet
adjusted assume augmented pattern belonging to category binary called Chapter cluster committee machine components Cornell Aeronautical Laboratory correction increment covariance matrix d-dimensional decision regions decision surfaces denote density function discussed dot products equal error-correction procedure Euclidean distance example Fix and Hodges fixed-increment error-correction function family g₁(X gi(X given hypersphere image-space implemented initial weight vectors layered machine linear dichotomies linear discriminant functions linearly separable loss function Lx(i mean vector minimum-distance classifier number of linear number of patterns optimum classifier parameters partition pattern classifier pattern hyperplane pattern points pattern space pattern vector pattern-classifying machines patterns belonging Perceptron piecewise linear point sets positive probability distributions prototype pattern PWL machine quadratic form quadric discriminant function quadric function sample covariance matrix solution weight vector Stanford subsets X1 Suppose training patterns training sequence training set training subsets values W₁ wa+1 weight point weight space X₁ X1 and X2 zero
Viitteet tähän teokseen
A Probabilistic Theory of Pattern Recognition Luc Devroye,László Györfi,Gabor Lugosi Rajoitettu esikatselu - 1997 |