Identifies the integral of the absolute Jacobian over an open source
unit ball with the volume of the closed target unit ball.
Statement
Let
,
put
,
and let
have its standard sup norm
and Lebesgue measure
.
Let
be continuous seminorms on
with specified positive constants
satisfying
The lower bounds make
and
norms. Define their unit spheres by
and
.
Let
be a bijection satisfying
Its radial extension, read as a map
in the fixed coordinates, is
For
,
the normalized vector
belongs to
,
so this expression is defined. Put
At a differentiability point
,
let
be the Fréchet derivative and set
At other points set
.
Thus
is the absolute Jacobian, not a norm of
or a determinant of the nonlinear map itself. Then
where the finite measure on the right is identified with its real
value.
Assumptions
The dimension is the positive integer
.
The maps defining
and
are continuous seminorms with the displayed positive comparison bounds.
The sphere map
is a bijective isometry for the distances induced by
and
.
The derivative and the measure in the integral use the fixed
coordinate space
,
whereas the unit spheres use
and
.
No linearity or everywhere differentiability of
is assumed.
Conclusion
The left side is the integral of the nonnegative function
over the open
-unit
ball. At a differentiability point, with coordinate vectors
and component functions
,
the integrand is explicitly
The right side is the volume of the closed
-unit
ball in the same coordinates. The theorem uses the absolute Jacobian and
does not assert that the radial extension is linear.
Changing from the norm
or
to the sup norm changes the normed-space structure, not the coordinate
vector. The reference Lebesgue normalization remains fixed. At a point
without a derivative, the source takes the zero linear map; since
,
its determinant is zero, exactly matching the definition of
above.
Proof route
Prove global Lipschitz control and the exact open-ball image. Apply
the nonnegative area formula after removing the null
nondifferentiability set, justify the real integral, and use the
zero-measure boundary of the target ball.
Proof steps
Control the map and identify its image. Radius
preservation and the inverse radial extension give
In particular,
is injective. The radial three-Lipschitz estimate and the comparison
bounds give
This is the
reference-coordinate Lipschitz bound for the same
.
Define the open target ball
.
The radius identity gives
.
Conversely, for
,
take
;
then
and
.
Hence
Apply the nonnegative area formula. Write
for the nonnegative extended integral, valued in
.
The formula is
Apply the
injective nonnegative area formula with measurable domain
,
map
,
derivative
at each
,
injectivity from Step 1, and reference Lebesgue measure
.
Step 2 then yields
The first equality uses
.
These are extended nonnegative integrals; real integrability is
established next.
Verify integrability of the same integrand. Put
.
This is compact by the comparison bounds. The integrability estimate is
Because
,
its finite extended integral is the same number as its real integral.
Taking real values in Step 3 therefore gives
with the finite measure on the right read as a real number.
Replace the open target ball by the closed ball.
The boundary identities are
The boundary description follows from the norm-ball description; its
measure is zero by the convex-set boundary theorem. Both steps are used
in the
open/closed target-volume lemma. Finiteness follows from compactness
of the closed
-unit
ball. Substitution into Step 4 gives the stated equality
The positive dimension
and the two stored seminorms
on the fixed sup-norm coordinate space
,
with the positive comparison bounds stated in this Card.
Space MX; Space MY
The same coordinate vectors equipped with norms
and
.
This changes the norm used for sphere distances, not the reference
coordinate measure.
(Δ : sphere (0 : Space MX) 1 ≃ᵢ sphere (0 : Space MY)
1)
The given bijective isometry
,
with
.
show Space MX from y
Regard the same vector
with norm
;
this is not an additional transformation.
radialExtension (X := Space MX) (Y := Space MY) Δ (show Space
MX from y)
The radial value
:
zero at zero, and
otherwise.
show Fin (m + 1) → ℝ from radialExtension (X := Space MX) (Y
:= Space MY) Δ (show Space MX from y)
Read that same value back in the fixed reference coordinates. The
entire function under the derivative is
.
ContinuousLinearMap.det (fderiv ℝ (fun y : Fin (m + 1) → ℝ
=> (show Fin (m + 1) → ℝ from radialExtension (X := Space MX) (Y :=
Space MY) Δ (show Space MX from y))) x)
The determinant
of the derivative of this whole coordinate map. The surrounding
absolute-value bars in the signature give
.
In the source
Mathematical meaning
MX.p.ball 0 1
The open source ball
.
∂volume = MY.closedUnitBallVolume
The complete equality is
,
where
.
The cited volume definition takes the real value of its finite measure.
The same reference Lebesgue measure is used throughout.