Week 11

Topic 2. Inverse Functions (Stewart, pp. 412 – 419)

Definition 1. A function f is called injective or one-to-one on a set S ⊆ Df when


∀a,b∈S f (a) = f (b) =⇒ a = b

Or stated differently, f is injective when

∀a,b∈S a ̸= b =⇒ f (a) ̸= f (b)

The notation “∀a,b∈S ” is just a compact way of writing “for all a and b in S”.
Theorem 2 (Horizontal Line Test). A function f is injective if and only if no horizontal line cuts
the graph of f more than once.
Illustration 3.

Example 4. Determine which of the following functions are injective on their domains.
√
(a) f (x) = x2 + 1 (b) g(x) = 2x + 1

Solution:
(a) Notice that (b) Set g(a) = g(b) vir a, b ∈ Dg . Then
√ √
f (−1) = (−1)2 + 1 = 2 2a + 1 = 2b + 1
2 =⇒ 2a + 1 = 2b + 1
f (1) = 1 + 1 = 2
=⇒ 2a = 2b
So f (−1) = f (1), but −1 ̸= 1.
=⇒ a=b
Thus, f is not injective.
Thus, g is injective.

Remark. Intuitively we can notice the following. If a function is continuous on an interval I, then
f is injective on I if and only if f is increasing or decreasing on I.
Since g in the previous example was continuous on its domain (an interval), we could have deduced
that g is injective on its domain. But for f , f is continuous on its domain (an interval), but f is

1
not increasing or decreasing everywhere on Df , and is thus not injective!

Definition 5. For an injective function f , its inverse function f −1 is defined as

∀y∈Wf , x∈Df f −1 (y) = x ⇐⇒ f (x) = y

Or equivalently, f ’s inverse function f −1 is the unique function with the properties

∀x∈Df f −1 f (x) = x and ∀y∈Wf f f −1 (y) = y

Further, the domain and range of the inverse function f −1 is

Df −1 = Wf and Wf −1 = Df

Remark. For a function f , if there is a function g such that ∀x∈Df gf (x) = x , then f is neccesarily
injective. As proof, take any a, b ∈ Df . Then

f (a) = f (b) =⇒ gf (a) = gf (b) =⇒ a=b

Consequently, the function for which such an inverse exists, are precisely the injective functions.

Illustration 6 (Graphs of inverse functions). For an injective function f , the definition of the
inverse function f −1 is also equivalent to

(a, b) lies on the graph of y = f (x) ⇐⇒ (b, a) lies on the graph of y = f −1 (x)

Geometrically, this means that the graphs of f and its inverse f −1 are symmetric around the
line y = x.

Theorem 7. To determine the inverse of a given injective function f , set f (y) = x, and solve
for y in terms of x. This then produces y = f −1 (x).

√
Example 8. Consider the function g(x) = 2x + 1
(a) Determine a formula for g −1 (x)
(b) What is the domain Dg−1 of g −1 ?

2
Solution:
√
(a) Let x = 2y + 1 and solve for y.

x2 = 2y + 1
=⇒ 2y = x2 − 1
1
x2 − 1 = g −1 (x)


=⇒ y= 2

(b) Dg−1 = Wg = [0, ∞)

x−1
Example 9. Let f (x) = . Determine a formula for f −1 (x), if f is injective.
4x + 7

Solution: Set f (a) = f (b), where a, b ∈ Df = R \ − 74 . Then



a−1 b−1
=
4a + 7 4b + 7
=⇒ (a − 1)(4b + 7) = (4a + 7)(b − 1)
=⇒ 4ab + 7a − 4b − 7 = 4ab + 7b − 4a − 7
=⇒ 11a = 11b =⇒ a=b

Thus, f is injective. Now, set x = f (y) and solve for y.

y−1
x= =⇒ 4xy + 7x = y − 1
4y + 7
7x + 1
=⇒ (4x − 1)y = −7x − 1 =⇒ y=− = f −1 (x)
4x − 1

Theorem 10. If f is an injective continuous function, then f −1 is also continuous.

Theorem 11. If f is an injective differentiable function, with inverse f −1 , then f −1 is also

differentiable and
d −1 1
f ′ f −1 (x) =


f (x) = ′ −1 , mits ̸ 0
dx f (f (x))

Prof: Suppose a ∈ Wf = Df −1 is a number such that f ′ f −1 (a) =

̸ 0. Then,

′ f −1 (x) − f −1 (a)
f −1 (a) = lim
x→a x−a

Let y = f −1 (x) and b = f −1 a

Then f (y) = x and f (b) = a

3
Further, since f −1 is continuous, lim y = lim f −1 (x) = f −1 (a) = b. Thus
x→a x→a
 
lim y = lim f −1 (x) = f −1 lim x = f −1 (a) = b
x→a x→a x→a

Thus

′ f −1 (x) − f −1 (a) y−b
f −1 (a) = lim = lim
x→a x−a y→b f (y) − f (b)
1 1 1
= = ′
= ′ −1
limy→b f (y)−f (b) f (b)
y−b
f (f (a))

which completes the proof.

Illustration 12. The result could also be intuitively observed from the following graphical example.

√ d −1 (x)
Example 13. Is f (x) = x3 + x + 9 injective? If so, determine dx x=3 f

Solution: Notice that f is continuous. Further

d
x3 + x + 9 = x2 + 1 > 0


dx
which is always strictly positive. Thus x3 + x + 9 is increasing with precisely one x-intercept a,
where x3 + x + 9 > 0 when x > a. Consequently Df = [a, ∞) is an interval on which f is continuous
and increasing, and thus f is injective.
√
Further, f (0) = 9 = 3 =⇒ f −1 (3) = 0
3x2 + 1
Also, f ′ (x) = √ =⇒ f ′ (0) = 1
2·3 = 1
6
2 x3 + x + 9
Thus
d 1
f −1 (x) = =6
dx x=3 f ′ (0)