Graphics Reference
In-Depth Information
(
)
h
xx
xx
,
1
1
Æ
0
as
xx
Æ
.
1
-
We must show that an equation similar to (4.8) holds for f -1 . The obvious candidate
for Df -1 is A = (Df) -1 . Applying A to both sides of equation (4.8) gives
(
) +
(
(
)
) =
-
1
() -
-
1
()
A
yy
-
A h
yy
,
f
y
f
y
,
1
1
1
1
where h 1 ( y , y 1 ) =-h(f -1 ( y ),f -1 ( y 1 )). Now
(
)
(
)
h
yy
yy
,
h
yy
xx
,
xx
yy
-
-
1
1
1
1
1
1
1
1
=-
-
-
and the right-hand side of this equation goes to zero (the first term goes to zero and
the second is bounded by 2). This shows that f -1 is differentiable. The continuity of
D(f -1 ) follows from the fact that the matrix for this map is defined by the composite
of the maps
r
Ê
Ë
ˆ
¯ æÆ
n
n
() æ
(
) æ
(
)
B
æ
B
r
æææææ
Æ
GL
n
,
R
ææææ æ
Æ
GL
n
,
R
,
2
-
1
matrix
for Df
matrix inversion
f
where we identify the space GL (n, R ) of nonsingular real n ¥ n matrices with R n 2 .
If f is of class C k , then one can show in a similar fashion that f -1 is also.
To prove the first application of the inverse function theorem we need two lemmas.
4.4.3. Lemma. Let U be an open subset of R n that contains the origin. Let f : U Æ
R m , n £ m, be a C k map with f( 0 ) = 0 and k ≥ 1. Assume that Df( 0 ) has rank n. Then
there is a C k diffeomorphism g of one neighborhood of the origin in R m onto another
with g( 0 ) = 0 and such that
(
(
)
) = (
)
g f x
,...,
x
x
,...,
x
, ,...,
00
1
n
1
n
holds in some neighborhood of the origin in R n .
Proof. The hypothesis that Df( 0 ) has rank n means that the n ¥ m Jacobian matrix
(∂f i /∂x j ) has rank n. Because we can interchange coordinates if necessary, there is no
loss in generality if we assume that
f
x
Ê
Á
ˆ
˜
i
rank
=
n
.
j
1,
££
ij n
Consider the map F : U ¥ R m-n
Æ R m defined by
(
) = (
) + (
)
Fx
,...,
x
fx
,...,
x
00
,..., ,
x
,...,
x
.
1
m
1
n
n
+
1
m
Since F(x 1 ,...,x n ,0,...,0) = f(x 1 ,...,x n ), F is an extension of f. Furthermore, the
determinant of (∂F i /∂x j ) is just the determinant of
 
Search WWH ::




Custom Search