Learning Machines: Foundations of Trainable Pattern-classifying SystemsMcGraw-Hill, 1965 - 137 sivua |
Kirjan sisältä
Tulokset 1 - 3 kokonaismäärästä 25
Sivu 81
... fixed - increment error - correction procedure and beginning with any initial weight vector W1 . Then , for some finite index ko , W ko = Wk = Wk ko + 1 = ko + 2 is a solution vector . Discussion The value of the fixed correction ...
... fixed - increment error - correction procedure and beginning with any initial weight vector W1 . Then , for some finite index ko , W ko = Wk = Wk ko + 1 = ko + 2 is a solution vector . Discussion The value of the fixed correction ...
Sivu 82
... fixed - increment error - correction weight - vector sequence Sw can now be produced recursively from the training sequence Sy by the simplified rule Wx + 1 = Wk Wk + 1 W + Yx ' = • if Yk ' Wk > 0 if Yk ' . Wk < 0 ( 5.8 ) where c is ...
... fixed - increment error - correction weight - vector sequence Sw can now be produced recursively from the training sequence Sy by the simplified rule Wx + 1 = Wk Wk + 1 W + Yx ' = • if Yk ' Wk > 0 if Yk ' . Wk < 0 ( 5.8 ) where c is ...
Sivu 87
... fixed - increment error - correction procedure , suitably modified by replacing zero by T in Eqs . ( 5 · 3 ) and ( 5 · 4 ) , is still guaranteed to produce a solution weight vector when one exists . 5.5 A training theorem for R ...
... fixed - increment error - correction procedure , suitably modified by replacing zero by T in Eqs . ( 5 · 3 ) and ( 5 · 4 ) , is still guaranteed to produce a solution weight vector when one exists . 5.5 A training theorem for R ...
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 apply assume bank called cells changes Chapter classifier cluster column committee machine components consider consists contains correction corresponding covariance decision surfaces define denote density depends described discriminant functions discussed distance distributions elements equal error-correction estimates example exist expression FIGURE fixed given implemented initial layered machine linear machine linearly separable lines majority matrix mean measurements modes negative networks nonparametric normal Note optimum origin parameters partition pattern hyperplane pattern space pattern vector pattern-classifying piecewise linear plane points positive presented probability problem properties PWL machine quadric regions respect response rule selection separable sequence side solution space Stanford step subsidiary discriminant Suppose theorem theory threshold training methods training patterns training procedure training sequence training subsets transformation values weight vectors X1 and X2 Y₁ zero
Viitteet tähän teokseen
A Probabilistic Theory of Pattern Recognition Luc Devroye,László Györfi,Gabor Lugosi Rajoitettu esikatselu - 1997 |