Implicit Differentiation

Explicit equation: z = F(x, y)

Implicit equation: F(x, y, z) = 0
𝜕𝜕𝜕𝜕 𝜕𝜕𝑥𝑥
(x and y are independent from each other, thus = 0, = 0)
𝜕𝜕𝜕𝜕 𝜕𝜕𝑦𝑦
From the chain rule, we have
𝜕𝜕𝑥𝑥 𝜕𝜕𝑦𝑦 𝜕𝜕𝑧𝑧
𝐹𝐹𝑥𝑥 + 𝐹𝐹𝑦𝑦 +𝐹𝐹𝑧𝑧 =0
𝜕𝜕𝑥𝑥 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕

𝜕𝜕𝑥𝑥 𝜕𝜕𝜕𝜕 𝜕𝜕𝑧𝑧

=1 =0 𝐹𝐹𝑥𝑥 +𝐹𝐹𝑧𝑧 =0
𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕
Likewise, differentiating w with respect to y leads us to

𝜕𝜕𝑥𝑥 𝜕𝜕𝑦𝑦 𝜕𝜕𝑧𝑧

𝐹𝐹𝑥𝑥 + 𝐹𝐹𝑦𝑦 +𝐹𝐹𝑧𝑧 =0
𝜕𝜕𝑦𝑦 𝜕𝜕𝑦𝑦 𝜕𝜕𝑦𝑦

𝜕𝜕𝑥𝑥 𝜕𝜕𝜕𝜕 𝜕𝜕𝑧𝑧

=0 =1 𝐹𝐹𝑦𝑦 +𝐹𝐹𝑧𝑧 =0
𝜕𝜕𝑦𝑦 𝜕𝜕𝑦𝑦 𝜕𝜕𝑦𝑦
Example 12.4.5.
Finding Partial Derivatives Implicitly
Solution:
 12.5
THE GRADIENT AND DIRECTIONAL
DERIVATIVES

5
Directional Derivative

 Sometimes, we need a way to find the rate of change

of a function in a given direction.

 For example, we want to find

the instantaneous rate of
change of f(x, y) at the point
P(a, b) and in the direction
given by the unit vector u = 〈u1,
u2〉.

 We can take directional derivatives.

6
Definition 12.5.1

The directional derivative of f(x, y) at the point (a, b)

and in the direction of the unit vector u = 〈u1, u2〉 is
given by

provided the limit exists.

• Notice that this limit resembles the definition of partial
derivative, except that in this case, both variables may
change.
• Further, observe that the directional derivative in the
direction of the positive x-axis (i.e., in the direction of the
unit vector u = 〈1, 0〉 is

• Likewise, the directional derivative in the direction of the

positive y-axis (i.e., in the direction of the unit vector u =
〈0, 1〉 is .
Theorem 12.5.1

Suppose that f is differentiable at (a, b) and u = 〈u1, u2〉

is any unit vector. Then, we can write
Example 12.5.1. Computing Directional
Derivatives

For compute for the

directions
(a) u = 〈 〉,
(b) u in the direction from (2, 1) to (4, 0).
Solution:

We have , so that

(a) u = 〈 〉
(b) u in the direction from (2, 1) to (4, 0).
We first find the unit vector u in the indicated direction.
The vector from (2, 1) to (4, 0) corresponds to the
position vector 〈2,−1〉.
So the unit vector in that direction is
Gradient

Definition 12. 5.2

The gradient of f(x, y) is the vector-valued function

provided both partial derivatives exist.

We denote the gradient of a function f by grad f or ∇f (read “del f ”).

Using the gradient, we can write a directional derivative as
the dot product of the gradient and the unit vector in the
direction of interest:

Theorem 12.5.2
If f is a differentiable function of x and y and u is any
unit vector, then
Example 12.5.2. Finding Directional Derivatives

For f(x, y) = x2 + y2, find Du f(1,−1) for

(a) u in the direction of v = 〈−3, 4〉,

(b) u in the direction of v = 〈3, −4〉.
Solution:

f(x, y) = x2 + y2

First, note that and

(a) u in the direction of v = 〈−3, 4〉

A unit vector in the same direction as v is ,

so
(b) u in the direction of v = 〈3, −4〉

A unit vector in the same direction as v is , so

Definition 12. 5. 3

The directional derivative of f(x, y, z) at the point

(a, b, c) and in the direction of the unit vector
u = 〈u1, u2, u3〉 is given by

The gradient of f(x, y, z) is the vector-valued function

provided all the partial derivatives are defined.

Second-Order Partial Derivatives
 For functions of two variables, there are four different second-
order partial derivatives.
 The partial derivative with respect to x of is
usually abbreviated as
∂f ∂ ∂f
 The partial derivative with respect to y of ∂y
is ( )
∂y ∂y
∂2 f
usually abbreviated as or fyy .
∂y 2
 For mixed second-order partial derivatives, one derivative is taken
with respect to each variable.

is abbreviated as

∂ ∂f ∂2 f
( ) is abbreviated as , or (fy)x = fyx.
∂x ∂y ∂x ∂y
 12.6
LOCAL EXTREMA OF FUNCTIONS OF
SEVERAL VARIABLES

20
Definition 12. 6. 1

We call f(a, b) a local maximum of f if there is an open

disk R centered at (a, b), for which f(a, b) ≥ f(x, y) for all
(x, y) ∈ R.
Definition 12. 6. 1

We call f(a, b) a local minimum of f if there is an open

disk R centered at (a, b), for which f(a, b) ≤ f(x, y) for all
(x, y) ∈ R.
Definition 12. 6. 2

The point (a, b) is a critical point of the function f(x, y)

if (a, b) is in the domain of f and either

or one or both of

do not exist at (a, b).

Theorem 12.6.1
If f(x, y) has a local extremum at (a, b), then (a, b) must
be a critical point of f.
Example 12.6.1. Finding Local Extrema
Graphically

Find all critical points of and

analyze each critical point graphically.
Solution:

(Since exponentials are always positive)

Since both partials exist for all (x, y), critical points are
solutions of

For this to occur, we must have , but x = 0

means ; there are no critical points with x = 0.

All combinations of and are critical points:

Critical points:

Keep in mind that the critical points are only candidates for
local extrema; we must look further to determine whether
they correspond to extrema.

Zoom in on each critical point in turn, to graphically

identify any local extrema:
 (1, 1)  local maximum

(-1, 1)  local minimum

The point (1, -1):
• In the plane x = 1
(extending left to right),
the point at (1, −1) is a
local minimum.

• In the plane y = −1 (extending back to front), the point

at (1, −1) is a local maximum.

• This point is therefore not a local extremum.

• (1, -1) is a saddle point of the surface.

The point (-1, -1):
• In the plane x = -1, the
point at (-1, −1) is a local
maximum.

• In the plane y = −1, the

point at (-1, −1) is a local
minimum.

• This point is therefore not a local extremum.

• (-1, -1) is a saddle point of the surface.

Definition 12. 6. 3

The point P(a, b, f(a, b)) is a saddle point of z = f(x, y) if

(a, b) is a critical point of f and if every open disk
centered at (a, b) contains points (x, y) in the domain
of f for which f(x, y) < f(a, b) and points (x, y) in the
domain of f for which f(x, y) > f(a, b).
Theorem 12.6.2
Suppose that f(x, y) has continuous second-order partial
derivatives in some open disk containing the point (a, b)
and that fx(a, b) = fy(a, b) = 0. Define the discriminant D
for the point (a, b) by

(i) If D(a, b) > 0 and fxx(a, b) > 0, then f has a local

minimum at (a, b).
(ii) If D(a, b) > 0 and fxx(a, b) < 0, then f has a local
maximum at (a, b).
(iii) If D(a, b) < 0, then f has a saddle point at (a, b).
(iv) If D(a, b) = 0, then no conclusion can be drawn.
Remark
To have D(a, b) > 0, we must have both fxx(a, b) > 0 and fyy(a, b) >
0 or both fxx(a, b) < 0 and fyy(a,b) < 0.
• In the first case, notice that the surface z = f(x, y) will be
concave up in the plane y = b and concave up in the plane x = a.
The surface will look like an upward-opening paraboloid near
the point (a,b). Consequently, f has a local minimum at (a, b).
• In the second case, both fxx(a, b) < 0 and fyy(a, b) < 0. This says
that the surface z = f(x, y) will be concave down in the
plane y = b and concave down in the plane x = a. So the surface
looks like a downward-opening paraboloid near the point (a, b)
and hence, f has a local maximum at (a, b).

Observe that one way to get D(a, b) < 0 is for fxx(a, b) and fyy(a, b)
to have opposite signs. To have opposite concavities in the
planes x = a and y = b means that there is a saddle point at (a, b).
Example 12.6.2. Using the Discriminant
to Find Local Extrema

Locate and classify all critical points for

f(x, y) = 3x2 − y3 − 6xy.
Solution:

f(x, y) = 3x2 − y3 − 6xy.

Both partials are defined for all (x, y), so we solve

The first equation gives y = x, which we substitute into

the second equation to get

So x = 0 or x = −2 and the corresponding y-values are y = 0

and y = −2, so that the only two critical points are (0, 0)
and (−2,−2).
 f(x, y) = 3x2 − y3 − 6xy.

Critical points are (0, 0) and (−2,−2).

To classify these points, we first compute the second
partial derivatives:
fxx = 6, fyy = −6y and fxy = −6

Then test the discriminant:

At (0, 0):
D(0, 0) < 0, there is a saddle point of f at (0, 0).
At (-2, -2):
D(−2,−2) > 0 and fxx(−2,−2) > 0, There is a local minimum
at (−2,−2).
Class Exercise

Locate and classify all critical points for

Solution:

exist for all (x, y), so the critical points are

solutions of the two equations

From the first equation we get

Substituting x = 0 into the second equation, we have

Substituting x = −2y into the second equation, we have

are critical points.

 are critical points.

To evaluate the discriminant D at each critical point:

At :

D < 0, f has a saddle point of f at .

At (-2, 1):
D(−2,1) > 0 and fxx(−2,1) = 6 (-2) + 6(1) = -6 < 0, f has a
local maximum at (−2,1).

At (0, 0):
D(0, 0) = 0, no conclusion can be drawn.
But in the plane y = 0 we have f(x, y) = x3.
In two dimensions, the curve z = x3 has an inflection
point at x = 0, so there is no local extremum at (0, 0).