Page 262 - Introduction to Statistical Pattern Recognition
P. 262
244 Introduction to Statistical Pattern Recognition
N?
af*
- Awj') A~i~)(Xi~)-fi~)~ ,-(Xl2)-h2)]do . (5.178)
!,=I 2x2
Using the expectations in (5.165)-(5.167), E, [ yy)} becomes
where an approximation of N-1 N is used to obtain the second line from the
first, Now, substituting the partial derivatives of (5.144) and (5.145) into
(5.179),
