MATH1021

Week 5 Lecture 2

Haotian Wu

The University of Sydney

2023 Semester 1
Lecture plan

Last time:
• Derivative
f (a+h) f (a)
- definition: f 0 (a) = lim h provided the limit exists
h!0

- geometric meaning: f 0 (a) is the slope of the tangent line to y = f (x) at x = a.

_Y=ti×
Today: t '"
,

• Di↵erentiation rules:
- basic rules
a

- chain rule
• Implicit di↵erentiation
- di↵erentiating inverse functions
- logarithmic di↵erentiation
• Mean Value Theorem
Higher order derivatives

• If f (x) is di↵erentiable, then the first order derivative is f 0 (x).

we

• If f 0 (x) is a a “nice” function, we can di↵erentiate it again (and again, ...).

• Notation: ④
3
✓ ◆ fix ) ✗ =

2nd order deriv 00 d 0 d d d2

.
f (x) = f (x) = f (x) = 2 f (x) '

f- ( , 3×2
dx dx dx dx × =

✓ ✓ ◆◆
d 00 d d d d3 "

f- Ix ) bx =

3rd order ' ' f 000 (x) = f (x) = f (x) = 3 f (x)

dx dx dx dx dx f ( )
'"
6 x =

.. " '
. f- ex , 0 =

d n
:
nth order is
f (n) (x) = n f (x) (di↵erentiate n times)
dx f (
" '

0
×, =

for all h
> 4
,
Di↵erentiation rules

You may freely use the following familiar results

(unless, of course, you are asked to prove them):

f (x) f 0 (x)
xn nxn 1

sin x cos x
cos x sin x
ex ex
1
ln x x
sinh x cosh x
cosh x sinh x
Di↵erentiation rules

• To di↵erentiate more complicated functions, we use various di↵erentiation rules.

• Theorem: If f (x) and g(x) are both di↵erentiable, then

Rnk :

1. (↵f )0 (x) = ↵f 0 (x) (↵ 2 R constant) ' '

I }
'
If g)
-
= f -
g
2. (f + g)0 (x) = f 0 (x) + g 0 (x)
.

' ' '

(f g)0 (x) = f 0 (x)g(x) + f (x)g 0 (x) If g) =/ f- g

3. (Product rule) wagging :

'

✓ ◆0 counterexample : ( 1.x ) = I
f f 0 (x)g(x) f (x)g 0 (x) *
4. (Quotient rule) (x) = (g(x) 6= 0)
g g(x)2 III. 1×11--0-1=0

5. (Power rule) (xa )0 = axa 1

(for any constant a 2 R)

6. (Chain rule) (f g)0 (x) = f 0 (g(x))g 0 (x)

Some di↵erentiation rules in other notation

• If u = f (x) and v = g(x), then

(product rule) (uv)0 = u0 v + uv 0

• If u = g(x) and y = f (g(x)) = f (u), then

Y
-

a-

&
dy dy du
(chain rule)
dx
=
du dx
=
thy)%u )
✗

• Remark: product chain + chain rule =) quotient rule

-2

¥ 1- 1) ✓ v1
' '

(E) 4- 1) u
'
"
+
Luv 1) ii. V + u
-
. .

= =
.

product chain rule

rule

=
wvjz . ""
ex Di↵erentiate (a) tan x (b) x ln(x3 )
u
✗
(a)
dalton ) =
£1"÷÷ ) ivj
Tv

=
Gs×ws×[,±n✗tsi
=
""j = ¥ = seix .

-1×1-4×1 = see ✗ .

"
"
 '
'

(b) ( ✗ lnlx )
?
) =
(x ) 'hl✗3 ) 1- ✗ .

Unix 's )

= t.hu )
}
+ ✗ .
.

3×2

= b. 1×31+3

= 3 lnx 1- 3

3×4×5
'

Or ,
( ✗ hix } ) ) = (
'

= 3 ( ✗ tax )

=3 ( I -

tax 1- ✗ .
)

=
34×+1 ) %
ex Prove the product rule (f g)0 = f 0 g + f g 0 by definition of derivative.
Proof:
-

0 (f g)(x + h) (f g)(x) g
(f g) (x) = lim
h!0 h
f (x + h)g(x + h) f (x)g(x)
= lim
h!0 h
f (x + h)g(x + h) f (x + h)g(x) + f (x + h)g(x) - f (x)g(x)
= lim mm m m mm

h!0
✓ h ◆
g(x + h) g(x) f (x + h) f (x)
h!0 ☐☐
= lim f (x + h) ·
h
+
☐ h
· g(x +
= h)

= f (x)g 0 (x) + f 0 (x)g(x).

- f dit
- -

P I
g differentiable ⇤
f differentiable fits him f- Ixth ) =
fix )
h→o
Implicit di↵erentiation

p
• An equation like y= 1 x2 is explicit.
• We can di↵erentiate both sides of this equation directly to find dy
dx :

dy d 1 x
= (1 x2 )1/2 = (1 x2 ) 1/2
( 2x) = x(1 x2 ) 1/2
= p
dx dx 2 1 x2

• Suppose x and y are related implicitly by the equation x2 + y 2 = 1.

dy
Question: Can we compute dx ? If so, how to compute?
Implicit di↵erentiation

• The equation x2 + y 2 = 1 does not define y as a function of x globally.

• However, “locally”, the equation implicitly defines y = f (x) as a function of x.1

^

locally around here ,

y =fÉ
iii.
\
, -

>

?
'
I

'
. .
'

locally around here

,
Y = -

1
The theory behind this is the Implicit Function Theorem, which will be discussed in MATH1023. We
don’t need all the details in MATH1021.
Implicit di↵erentiation

• In the example x2 + y 2 = 1, we can solve the implicit equation for y to get y = y(x),
dy
and then calculate dx = y 0 (x).

• x2 + y 2 = 1 + x cos(2x) sin(3y) implicitly defines y = y(x) around (x, y) = (0, 1).

1.5

1.0

0.5

0.0

easy !
-0.5

not
-1.0

I
-1.5

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

Can we solve for y to get y = y(x) near (0, 1)?

Can we still compute y 0 (0)? ← Sure we can !
Implicit di↵erentiation

• Suppose x and y are defined implicitly by a relation of the form

F (x, y) = 0.

• It is possible to calculate the derivative dy

dx using implicit di↵erentiation without
solving for y = y(x) explicitly.

• Implicit di↵erentiation consists of

1. di↵erentiating both sides of F (x, y) = 0 with respect to x,
2. remembering that y = y(x) is a function of x, I apply chain rule )
dy
3. then solve for dx .
ex x2 + y 2 = 1 + x cos(2x) sin(3y) implicitly defines y = y(x) around (x, y) = (0, 1).
dy
(a) Compute dx .
(b) Find the equation of the tangent line to the curve at (x, y) = (0, 1)
""
50¥ £ ¥ ✗+ )
?

1×2+1 )
:
= ✗ cos ( 2x ) Sinlsy )

2×+27
-

¥ = O t 1 .

cos th ) sin by )

1- ✗ .

I -
sink ) .
2) Sino y)

+ ✗ .

cos (4) .

cos ↳y) .

3¥×

.
-
.
 Cos ( 2x ) sin ( by ) -

2x sink ) Sindy ) -
2x
'

Y =

ay -

> ✗ Cosin ) GUY )

(b) At ✗ =0
,
4=1 :

"
sins
✗ ( o) = .
.
.
=
-2

'

-
.
9-4 of tangent line at 10,1 ) is

Y -

I =
5¥ (✗ - o ) Y =
si¥ ✗ + I
%
Di↵erentiating inverse functions

• This is just an application of implicit di↵erentiation.

• Suppose y = f 1 (x). Then f (y) = x.

Then by implicit di↵erentiation and chain rule,

f 0 (y)y 0 = 1,

so then
0 1
y0 = f 1
(x) = .
f 0 (f 1 (x))
T
 d 1
ex Compute dx sinh x.

"
Sol Y = Sinh ✗

✗ = Sinh y
( da both sides )

I Y
' the costly -
sink 'y= ,
=
cushy

ÉÉ¥×ÑÑ
.

'
Y =

£× ×_
i. "
Sinh ✗ =
. ""
Logarithmic di↵erentiation

• This is just another application of implicit di↵erentiation.

• Suppose y = f (x)g(x) . Then

✓ ◆
0 g 0 gf 0
y =f g ln f +
f

I try =
sent ÷
,
y
'
= silent -18¥
£ both sides

-1M¥ )
>
i. Y
'
=
Yfiht -18¥ ) =
f Is 'hf ☒
ex Let y = xx . Compute y 0 .

⑨ by =
h☒× )
£ both sides
hey = ✗ h✗

f-
'
Y =
1. hx + ✗ -

'
Y =
Ylhnxti )

' ✗
i , y =
✗ ltnxtl ) ""
.
ex Let g = (cos x)sin x . Compute g 0 .

hxercise

( )
'
"" sinxtanx
Answer : G =
cos ✗ Gsx lnlcosx ) -

.
Mean Value Theorem (MVT)2

• Theorem: If f is di↵erentiable on some interval I containing the points a and b.

Then there exists a number c 2 (a, b) such that

f (b) f (a)
f 0 (c) = .
b a

• Remarks:
I
- MVT guarantees the existence of at least one c 2 (a, b) s.t.
f 0 (c) = f (b)b fa (a) .
1
1
I
I 1
I

- MVT tells you such c exists, but does not tell you the exact
I
1

:c :

value of c.

2
Not examinable in MATH1021. This Theorem will be used in Week 7 (proof of the Lagrange form of
the remainder of Taylor series) and Week 10 (proof of the Fundamental Theorem of Calculus).
Consequence of the Mean Value Theorem

• The Mean Value Theorem has many important and useful consequences. Here, we
just state and prove one of them.

⇐ )
tra
• Theorem: If f 0 (x) = 0 for all x 2 (a, b), then f is constant on (a, b).
reading
Proof: Pick any x1 , x2 2 (a, b). Without loss of generality, we can assume x1 < x2 .
Since f is di↵erentiable on (a, b) (x1 , x2 ), f is di↵erentiable on (x1 , x2 ). Then by
the Mean Value Theorem, there exists c 2 (x1 , x2 ) ⇢ (a, b) such that

f (x2 ) f (x1 )
f 0 (c) = () f (x2 ) f (x1 ) = f 0 (c)(x2 x1 ) = 0
x2 x1

since f 0 (c) = 0 by the hypothesis that f 0 (x) = 0 for all x 2 (a, b).
Therefore, f (x1 ) = f (x2 ) for any x1 , x2 2 (a, b). That is, f is constant on (a, b). ⇤
Lecture summary

After today’s lecture, you should be able to:

• Di↵erentiate familiar functions
• Apply di↵erentiation rules to calculate derivatives of more complicated functions
• Apply implicit di↵erentiation to calculate derivatives
(better to remember the process rather than the formula)
• Understand the Mean Value Theorem