Skip to content

Unit 3: Differentiation: Composite, Implicit, and Inverse Differentiation

AP Calc cheatsheet

Open β†—

Loading…

Theorem (Chain Rule). If y=f(g(x))y = f(g(x)), then

dydx=fβ€²(g(x))gβ€²(x).\frac{dy}{dx} = f'(g(x))g'(x).

The outside derivative is evaluated at the inside function, and then multiplied by the derivative of the inside function.

Proof (Chain rule). Let u=g(x)u=g(x) and y=f(u)y=f(u). A small change in xx creates a small change in uu, and that creates a small change in yy. The ratio can be split as

Ξ”yΞ”x=Ξ”yΞ”uβ‹…Ξ”uΞ”x.\frac{\Delta y}{\Delta x} = \frac{\Delta y}{\Delta u}\cdot\frac{\Delta u}{\Delta x}.

As Δx→0\Delta x\to0, the intermediate change Δu→0\Delta u\to0 for a differentiable inside function. The two ratios approach f′(u)f'(u) and g′(x)g'(x), so

dydx=fβ€²(g(x))gβ€²(x).\frac{dy}{dx}=f'(g(x))g'(x).

The chain rule is about layered change. If xx changes, then the inside function g(x)g(x) changes first. That change then causes the outside function f(g(x))f(g(x)) to change.

Symbolically,

dydx=dyduβ‹…dudx.\frac{dy}{dx} = \frac{dy}{du}\cdot\frac{du}{dx}.

The notation makes the chain rule look like canceling fractions. That is a useful way to remember it, but derivatives are limits of ratios, so algebra with derivative notation still needs a rule that justifies it.

AP problems often hide the chain rule inside:

  • powers of expressions,
  • trig functions with nontrivial angles,
  • exponentials with nontrivial exponents,
  • logarithms of expressions,
  • inverse trig functions with expressions inside.

Whenever you differentiate an outside function, pause and ask what the inside function is.

Example. Differentiate y=(3x2+1)5y=(3x^2+1)^5.

The outside function is a fifth power and the inside function is 3x2+13x^2+1. Differentiate the outside power, keeping the inside unchanged, then multiply by the derivative of the inside:

dydx=5(3x2+1)4β‹…ddx(3x2+1).\frac{dy}{dx}=5(3x^2+1)^4\cdot\frac{d}{dx}(3x^2+1).

Since ddx(3x2+1)=6x\frac{d}{dx}(3x^2+1)=6x,

dydx=5(3x2+1)4(6x)=30x(3x2+1)4.\frac{dy}{dx}=5(3x^2+1)^4(6x)=30x(3x^2+1)^4.

Note that the chain rule should be used for all layers of a function. For example, if you have a function f(g(h(x)))f(g(h(x))), the derivative is equal to:

ddxf(g(h(x)))=fβ€²(g(h(x)))gβ€²(h(x))hβ€²(x).\frac{d}{dx} f(g(h(x))) = f'(g(h(x)))g'(h(x))h'(x).

This is correct because each layer contributes its own derivative factor, starting from the outside layer and moving inward.

For a composition with any number of layers,

F(x)=f1(f2(β‹―fn(x)⋯ )),F(x)=f_1(f_2(\cdots f_n(x)\cdots)),

differentiate each layer once, from the outside toward the inside:

Fβ€²(x)=f1β€²(f2(β‹―fn(x)⋯ ))f2β€²(f3(β‹―fn(x)⋯ ))β‹―fnβ€²(x).F'(x) = f_1'(f_2(\cdots f_n(x)\cdots)) f_2'(f_3(\cdots f_n(x)\cdots)) \cdots f_n'(x).

Every derivative factor is evaluated at the expression immediately inside its layer. The process stops only when the derivative reaches xx.

Example. Differentiate

y=esin⁑((3x2βˆ’1)4).y=e^{\sin((3x^2-1)^4)}.

There are four changing layers: the exponential, sine, fourth power, and quadratic. Differentiate them in that order:

dydx=esin⁑((3x2βˆ’1)4)cos⁑((3x2βˆ’1)4)β‹…4(3x2βˆ’1)3β‹…6x.\frac{dy}{dx} = e^{\sin((3x^2-1)^4)} \cos((3x^2-1)^4) \cdot4(3x^2-1)^3 \cdot6x.

Therefore,

dydx=24x(3x2βˆ’1)3cos⁑((3x2βˆ’1)4)esin⁑((3x2βˆ’1)4).\frac{dy}{dx} = 24x(3x^2-1)^3 \cos((3x^2-1)^4) e^{\sin((3x^2-1)^4)}.

Example. Let

F(x)=(x2+1)e(xβˆ’1)2x+1.F(x)=\frac{(x^2+1)e^{(x-1)^2}}{x+1}.

Find the equation of the tangent line to FF at x=2x=2.

The numerator is a product containing a composite exponential. Let

N(x)=(x2+1)e(xβˆ’1)2.N(x)=(x^2+1)e^{(x-1)^2}.

Use the product and chain rules:

Nβ€²(x)=2xe(xβˆ’1)2+(x2+1)e(xβˆ’1)2β‹…2(xβˆ’1).N'(x) = 2xe^{(x-1)^2} +(x^2+1)e^{(x-1)^2}\cdot2(x-1).

Now apply the quotient rule:

Fβ€²(x)=Nβ€²(x)(x+1)βˆ’N(x)(x+1)2.F'(x) = \frac{N'(x)(x+1)-N(x)}{(x+1)^2}.

At x=2x=2,

N(2)=5eandNβ€²(2)=4e+10e=14e.N(2)=5e \qquad\text{and}\qquad N'(2)=4e+10e=14e.

Thus

F(2)=5e3F(2)=\frac{5e}{3}

and

Fβ€²(2)=(14e)(3)βˆ’5e9=37e9.F'(2) = \frac{(14e)(3)-5e}{9} = \frac{37e}{9}.

The tangent line is

yβˆ’5e3=37e9(xβˆ’2).y-\frac{5e}{3} = \frac{37e}{9}(x-2).

Implicit equations describe a relationship rather than a solved function, meaning that the xxs and yys of the equation are not fully separated. To take dydx\frac{dy}{dx} of such function we can use implicit differentiation.

After differentiating, collect every term containing dy/dxdy/dx on one side. Factor out dy/dxdy/dx only after all product and chain rules have been expanded. The remaining coefficient determines the slope. If that coefficient is zero while the other side is nonzero, the curve may have a vertical tangent rather than an ordinary finite slope.

Example. Find the slope of the tangent line of the circle x2+y2=25x^2+y^2=25 at the point (3,4)(3,4).

Β‘535Β‘545(3;4)tangentslopeΒ‘xyxy

Differentiate both sides with respect to xx, remembering that yy depends on xx:

2x+2ydydx=0.2x+2y\frac{dy}{dx}=0.

Solve for dydx\frac{dy}{dx}:

dydx=βˆ’xy.\frac{dy}{dx}=-\frac{x}{y}.

Now substitute the point (3,4)(3,4):

dydx∣(3,4)=βˆ’34.\frac{dy}{dx}\Big|_{(3,4)}=-\frac{3}{4}.

So the tangent line at (3,4)(3,4) has slope βˆ’34-\frac34.

Example. The curve

x2y+sin⁑(xy)=x+sin⁑1x^2y+\sin(xy)=x+\sin1

passes through (1,1)(1,1). Find the equation of its tangent line at that point.

Differentiate both sides with respect to xx. The term x2yx^2y needs the product rule, while sin⁑(xy)\sin(xy) needs both the chain rule and another product rule:

2xy+x2dydx+cos⁑(xy)(xdydx+y)=1.2xy+x^2\frac{dy}{dx} +\cos(xy)\left(x\frac{dy}{dx}+y\right) =1.

Collect the derivative terms:

(x2+xcos⁑(xy))dydx=1βˆ’2xyβˆ’ycos⁑(xy).\left(x^2+x\cos(xy)\right)\frac{dy}{dx} = 1-2xy-y\cos(xy).

Therefore,

dydx=1βˆ’2xyβˆ’ycos⁑(xy)x2+xcos⁑(xy).\frac{dy}{dx} = \frac{1-2xy-y\cos(xy)}{x^2+x\cos(xy)}.

At (1,1)(1,1),

dydx=βˆ’1βˆ’cos⁑11+cos⁑1=βˆ’1.\frac{dy}{dx} = \frac{-1-\cos1}{1+\cos1} =-1.

The tangent line is

yβˆ’1=βˆ’(xβˆ’1),y-1=-(x-1),

or

y=βˆ’x+2.y=-x+2.

Theorem (Inverse Function Derivative). If ff is differentiable and invertible with fβ€²(a)β‰ 0f'(a) \ne 0, then

(fβˆ’1)β€²(b)=1fβ€²(a)(f^{-1})'(b) = \frac{1}{f'(a)}

where b=f(a)b = f(a).

Equivalent formula:

(fβˆ’1)β€²(x)=1fβ€²(fβˆ’1(x)).(f^{-1})'(x) = \frac{1}{f'(f^{-1}(x))}.

The derivative of an inverse is a reciprocal slope, but the reciprocal is taken at the matching point on the original function. If f(a)=bf(a)=b, then the point (a,b)(a,b) on ff becomes (b,a)(b,a) on fβˆ’1f^{-1}. The slope fβ€²(a)f'(a) belongs to the original point, while (fβˆ’1)β€²(b)(f^{-1})'(b) belongs to the reflected point.

The condition fβ€²(a)β‰ 0f'(a)\ne0 matters. A horizontal tangent on ff reflects to a vertical tangent on its inverse, so the inverse does not have a finite derivative there.

Proof (Inverse Function Derivative). If y=f(x)y=f(x) and x=fβˆ’1(y)x=f^{-1}(y), then composing the functions gives

f(fβˆ’1(x))=x.f(f^{-1}(x))=x.

Differentiate both sides:

fβ€²(fβˆ’1(x))(fβˆ’1)β€²(x)=1.f'(f^{-1}(x))(f^{-1})'(x)=1.

Solving for the inverse derivative gives

(fβˆ’1)β€²(x)=1fβ€²(fβˆ’1(x)).(f^{-1})'(x)=\frac{1}{f'(f^{-1}(x))}.

Example. Let f(x)=x3+xf(x)=x^3+x. Given that f(1)=2f(1)=2, find (fβˆ’1)β€²(2)(f^{-1})'(2).

Here a=1a=1 and b=2b=2, since f(1)=13+1=2f(1)=1^3+1=2. First compute the derivative of ff:

fβ€²(x)=3x2+1.f'(x)=3x^2+1.

Evaluate it at a=1a=1:

fβ€²(1)=3(1)2+1=4.f'(1)=3(1)^2+1=4.

By the inverse function rule,

(fβˆ’1)β€²(2)=1fβ€²(1)=14.(f^{-1})'(2)=\frac{1}{f'(1)}=\frac{1}{4}.

Example. Suppose f(2)=5f(2)=5 and fβ€²(2)=βˆ’3f'(2)=-3. Define

g(x)=fβˆ’1(x2+1).g(x)=f^{-1}(x^2+1).

Find the equation of the tangent line to gg at x=2x=2.

First find the point on gg:

g(2)=fβˆ’1(5)=2.g(2)=f^{-1}(5)=2.

Differentiate the composite inverse function:

gβ€²(x)=(fβˆ’1)β€²(x2+1)β‹…2x=2xfβ€²(fβˆ’1(x2+1)).g'(x) = (f^{-1})'(x^2+1)\cdot2x = \frac{2x}{f'(f^{-1}(x^2+1))}.

At x=2x=2,

gβ€²(2)=4fβ€²(fβˆ’1(5))=4fβ€²(2)=βˆ’43.g'(2) = \frac4{f'(f^{-1}(5))} = \frac4{f'(2)} = -\frac43.

Using the point (2,2)(2,2), the tangent line is

yβˆ’2=βˆ’43(xβˆ’2).y-2=-\frac43(x-2).

ddx(arcsin⁑x)=11βˆ’x2\frac{d}{dx}(\arcsin x) = \frac{1}{\sqrt{1-x^2}} ddx(arccos⁑x)=βˆ’11βˆ’x2\frac{d}{dx}(\arccos x) = -\frac{1}{\sqrt{1-x^2}} ddx(arctan⁑x)=11+x2\frac{d}{dx}(\arctan x) = \frac{1}{1+x^2} ddx(arcsec⁑x)=1∣x∣x2βˆ’1\frac{d}{dx}(\operatorname{arcsec} x) = \frac{1}{\lvert x\rvert\sqrt{x^2-1}} ddx(arccsc⁑x)=βˆ’1∣x∣x2βˆ’1\frac{d}{dx}(\operatorname{arccsc} x) = -\frac{1}{\lvert x\rvert\sqrt{x^2-1}} ddx(arccot⁑x)=βˆ’11+x2\frac{d}{dx}(\operatorname{arccot} x) = -\frac{1}{1+x^2}

For AP work, arcsin⁑\arcsin, arccos⁑\arccos, and arctan⁑\arctan are the most common inverse trig functions. All six formulas follow from the inverse-function derivative formula; one proof is shown below.

The listed formulas give the derivative of the inverse trig function itself. If its input is another function u(x)u(x), the chain rule adds a factor of uβ€²(x)u'(x). For example,

ddxarcsin⁑(u)=uβ€²1βˆ’u2,\frac{d}{dx}\arcsin(u) = \frac{u'}{\sqrt{1-u^2}},

and

ddxarctan⁑(u)=uβ€²1+u2.\frac{d}{dx}\arctan(u) = \frac{u'}{1+u^2}.

Domain restrictions still matter after differentiating. The derivative of arcsin⁑x\arcsin x becomes unbounded at x=±1x=\pm1, and the absolute value in the arcsec and arccsc formulas cannot be dropped without knowing the sign of the input.

Proof (Derivative of arcsin⁑x\arcsin x).

Method 1 (Implicit Differentiation). Let y=arcsin⁑x.y=\arcsin x.

This means

sin⁑y=x,βˆ’Ο€2≀y≀π2.\sin y=x, \qquad -\frac{\pi}{2}\le y\le \frac{\pi}{2}.

Differentiate implicitly with respect to xx:

cos⁑ydydx=1.\cos y\frac{dy}{dx}=1.

So

dydx=1cos⁑y.\frac{dy}{dx}=\frac{1}{\cos y}.

Since sin⁑y=x\sin y=x and yy is in the principal range of arcsine, cos⁑yβ‰₯0\cos y\ge0. Using sin⁑2y+cos⁑2y=1\sin^2 y+\cos^2 y=1,

cos⁑y=1βˆ’sin⁑2y=1βˆ’x2.\cos y=\sqrt{1-\sin^2 y}=\sqrt{1-x^2}.

Therefore

ddxarcsin⁑x=11βˆ’x2.\frac{d}{dx}\arcsin x=\frac{1}{\sqrt{1-x^2}}.

Method 2 (Inverse Function Formula). Let f(y)=sin⁑yf(y)=\sin y, so fβˆ’1(x)=arcsin⁑xf^{-1}(x)=\arcsin x. The inverse derivative formula gives

(fβˆ’1)β€²(x)=1fβ€²(fβˆ’1(x)).(f^{-1})'(x)=\frac{1}{f'(f^{-1}(x))}.

Since fβ€²(y)=cos⁑yf'(y)=\cos y,

ddxarcsin⁑x=1cos⁑(arcsin⁑x).\frac{d}{dx}\arcsin x = \frac{1}{\cos(\arcsin x)}.

If ΞΈ=arcsin⁑x\theta=\arcsin x, then sin⁑θ=x\sin\theta=x and ΞΈ\theta is in the principal range where cos⁑θβ‰₯0\cos\theta\ge0. Therefore

cos⁑(arcsin⁑x)=1βˆ’x2,\cos(\arcsin x)=\sqrt{1-x^2},

so

ddxarcsin⁑x=11βˆ’x2.\frac{d}{dx}\arcsin x=\frac{1}{\sqrt{1-x^2}}.

Example. Differentiate y=arctan⁑(x2)y=\arctan(x^2).

Use the chain-rule form with inside function u=x2u=x^2, so uβ€²=2xu'=2x:

dydx=uβ€²1+u2=2x1+(x2)2.\frac{dy}{dx}=\frac{u'}{1+u^2}=\frac{2x}{1+(x^2)^2}.

Simplifying the square gives

dydx=2x1+x4.\frac{dy}{dx}=\frac{2x}{1+x^4}.

Example. Differentiate and simplify

y=arcsin⁑(x1+x2).y=\arcsin\left(\frac{x}{\sqrt{1+x^2}}\right).

Let

u=x1+x2=x(1+x2)βˆ’1/2.u=\frac{x}{\sqrt{1+x^2}} = x(1+x^2)^{-1/2}.

Differentiate uu using the product and chain rules:

uβ€²=(1+x2)βˆ’1/2βˆ’x2(1+x2)βˆ’3/2=1(1+x2)3/2.u' = (1+x^2)^{-1/2} -x^2(1+x^2)^{-3/2} = \frac1{(1+x^2)^{3/2}}.

Also,

1βˆ’u2=1βˆ’x21+x2=11+x2.1-u^2 = 1-\frac{x^2}{1+x^2} = \frac1{1+x^2}.

Since 1+x2>01+x^2>0,

1βˆ’u2=11+x2.\sqrt{1-u^2}=\frac1{\sqrt{1+x^2}}.

Apply the chain-rule form of the arcsine derivative:

dydx=uβ€²1βˆ’u2=1/(1+x2)3/21/1+x2=11+x2.\frac{dy}{dx} = \frac{u'}{\sqrt{1-u^2}} = \frac{1/(1+x^2)^{3/2}}{1/\sqrt{1+x^2}} = \frac1{1+x^2}.

Logarithmic differentiation is useful when a function has products, quotients, powers, or variables in both the base and exponent. The idea is to take the natural logarithm of both sides, use log laws to simplify, and then differentiate implicitly.

For example, if

y=(x2+1)3xβˆ’4,y=(x^2+1)^3\sqrt{x-4},

then taking logs gives

ln⁑y=3ln⁑(x2+1)+12ln⁑(xβˆ’4).\ln y=3\ln(x^2+1)+\frac12\ln(x-4).

Differentiate both sides:

1ydydx=3β‹…2xx2+1+12β‹…1xβˆ’4.\frac{1}{y}\frac{dy}{dx} =3\cdot\frac{2x}{x^2+1}+\frac12\cdot\frac{1}{x-4}.

Then multiply by yy:

dydx=(x2+1)3xβˆ’4(6xx2+1+12(xβˆ’4)).\frac{dy}{dx} =(x^2+1)^3\sqrt{x-4} \left(\frac{6x}{x^2+1}+\frac{1}{2(x-4)}\right).

Logarithmic differentiation is a strategy, not a new derivative rule. It works because logarithms turn complicated multiplication, division, and powers into simpler operations:

ln⁑(ab)=ln⁑a+ln⁑b,\ln(ab)=\ln a+\ln b, ln⁑(ab)=ln⁑aβˆ’ln⁑b,\ln\left(\frac{a}{b}\right)=\ln a-\ln b, ln⁑(ar)=rln⁑a.\ln(a^r)=r\ln a.

This is especially helpful when a function has many factors or when a variable appears in both the base and the exponent.

For a variable power u(x)v(x)u(x)^{v(x)} with u(x)>0u(x)>0, logarithmic differentiation turns the exponent into a factor:

ln⁑y=v(x)ln⁑(u(x)).\ln y=v(x)\ln(u(x)).

Differentiating gives

yβ€²y=vβ€²(x)ln⁑(u(x))+v(x)uβ€²(x)u(x).\frac{y'}{y} = v'(x)\ln(u(x)) +v(x)\frac{u'(x)}{u(x)}.

This contains both the derivative of the exponent and the derivative of the base. Treating v(x)v(x) as a constant would miss the first term.

After differentiating, remember that differentiating ln⁑y\ln y gives

1ydydx,\frac{1}{y}\frac{dy}{dx},

so the final derivative usually comes from multiplying by yy, where you resubstitude the original function (if y=f(x)y=f(x)).

Example. Use logarithmic differentiation to find dydx\frac{dy}{dx} for y=xxy=x^x (with x>0x>0).

The variable appears in both the base and the exponent, so take the natural logarithm of both sides:

ln⁑y=ln⁑(xx)=xln⁑x.\ln y=\ln\left(x^x\right)=x\ln x.

Differentiate both sides with respect to xx. The left side uses the chain rule, and the right side uses the product rule:

1ydydx=ln⁑x+xβ‹…1x=ln⁑x+1.\frac{1}{y}\frac{dy}{dx}=\ln x+x\cdot\frac{1}{x}=\ln x+1.

Multiply both sides by yy and substitute y=xxy=x^x:

dydx=xx(ln⁑x+1).\frac{dy}{dx}=x^x(\ln x+1).

Example. Use logarithmic differentiation to find dy/dxdy/dx for

y=(x2+1)sin⁑xx3xβˆ’1,x>1.y=\frac{(x^2+1)^{\sin x}}{x^3\sqrt{x-1}}, \qquad x>1.

Take the natural logarithm and expand using log laws:

ln⁑y=sin⁑xln⁑(x2+1)βˆ’3ln⁑xβˆ’12ln⁑(xβˆ’1).\ln y = \sin x\ln(x^2+1) -3\ln x -\frac12\ln(x-1).

Differentiate both sides. The first term requires the product rule:

1ydydx=cos⁑xln⁑(x2+1)+2xsin⁑xx2+1βˆ’3xβˆ’12(xβˆ’1).\frac1y\frac{dy}{dx} = \cos x\ln(x^2+1) +\frac{2x\sin x}{x^2+1} -\frac3x -\frac1{2(x-1)}.

Multiply by yy and substitute the original expression:

dydx=(x2+1)sin⁑xx3xβˆ’1(cos⁑xln⁑(x2+1)+2xsin⁑xx2+1βˆ’3xβˆ’12(xβˆ’1)).\frac{dy}{dx} = \frac{(x^2+1)^{\sin x}}{x^3\sqrt{x-1}} \left( \cos x\ln(x^2+1) +\frac{2x\sin x}{x^2+1} -\frac3x -\frac1{2(x-1)} \right).

This unit is about recognizing when the derivative is hidden inside another relationship.

  • Chain rule problems hide a changing input inside an outer function.
  • Implicit differentiation hides yy as a function of xx.
  • Inverse-function problems hide the slope relationship between a function and its inverse.
  • Related rates hide time dependence inside geometry or context.

Before moving on, make sure you can explain which rule is being used at each step. On AP-style questions, the hardest part is often choosing the rule order, not doing the algebra afterward.