Composite · Implicit
& Inverse Derivatives
The chain rule unlocks every nested function you've ever seen — sin(x²), √(1 + cos t), (3x+1)5. Implicit differentiation lets you find slopes on curves you can't even solve for y. Inverse derivatives finish the toolkit.
By the end of this unit, you will be able to…
- LO 1 State and apply the chain rule to differentiate composite functions in any form.
- LO 2 Differentiate implicitly-defined functions and find slopes of tangent lines on closed-form curves (circles, ellipses, etc.).
- LO 3 Compute derivatives of inverse functions using the inverse function theorem.
- LO 4 Differentiate arcsin, arccos, arctan, arccot, arcsec, arccsc and combinations with the chain rule.
- LO 5 Set up and solve related-rates problems using chain rule (foundation for Unit 4).
- LO 6 Justify the chain rule using the limit definition and explain each step in a free response.
Unit 3 → AP Topic reference
Unit 3 pulls skills across the 2024-25 CED Topic 3 (Differentiation) and Topic 4 (Contextual Applications).
| Sub-topic | CED Skill ID | What it covers |
|---|---|---|
| 3.1 Composite functions & chain rule | FUN-3.C | differentiate f∘g, f∘g∘h, table form |
| 3.2 Implicit differentiation | FUN-3.D | tangent lines on closed curves, second derivatives |
| 3.3 Inverse function theorem | FUN-3.E | computing (f⁻¹)'(b) from f' |
| 3.4 Derivatives of inverse trig | FUN-3.F | arcsin, arccos, arctan, arccot, arcsec, arccsc |
| 3.5 Differentiation strategy | FUN-3.A/B/C/D/E/F | picking the right rule in any combination |
Key terms to know before slide 5
Composite function
(f∘g)(x) = f(g(x)) — apply g, then f to the result. Order matters.Inner function
The "g(x)" inside f — the one you evaluate first.Outer function
The "f" applied last — its derivative gets multiplied by the inner derivative.Chain rule
(f∘g)'(x) = f'(g(x)) · g'(x). Multiply derivatives, evaluate.Implicit relation
Equation defining y as a function of x without solving for y. e.g., x² + y² = 25.Inverse function
f⁻¹ undoes f: f⁻¹(f(x)) = x. Reflection across y = x.Notation for the same thing, multiple ways
| Notation | Used for |
|---|---|
| d/dx [f(g(x))] | Chain rule in operational form |
| f'(g(x)) · g'(x) | Lagrange style — same thing, expanded |
| (f∘g)'(x) | When the composite has a name |
| dy/du · du/dx | Leibniz "cancelling" style — most intuitive for related rates |
| d/dx [y] = dy/dx | Implicit: differentiate both sides |
| (f⁻¹)'(b) | Derivative of inverse at point b; equals 1/f'(a) where b = f(a) |
Why "outer times inner" works
Take the limit as Δx → 0: (f∘g)'(x) = f'(g(x)) · g'(x). The intermediate variable "cancels" if you use Leibniz.
The chain rule, formally · FUN-3.C.1
| Type | Pattern |
|---|---|
| Polynomial inside polynomial | d/dx [(3x² + 1)5] = 5(3x² + 1)4 · 6x |
| Trig inside polynomial | d/dx [sin(πx²)] = cos(πx²) · 2πx |
| Exp inside polynomial | d/dx [e5x+1] = e5x+1 · 5 |
| Any nested combination | differentiate outer (treat inside as a variable) · multiply by derivative of inside |
Basic chain rule
Chain rule with trig
Triple-nested functions
Chain rule in a table (when there are 3+ layers)
| Function | sin | tan | · | x² |
|---|---|---|---|---|
| Derivative rule | cos(u) | sec²(v) | 2x | |
| Substitute | cos(tan(x²)) | sec²(x²) | · | 2x |
| Multiply | 2x · sec²(x²) · cos(tan(x²)) | |||
Chain rule with ex and ln x
When you can't (or won't) solve for y
Find the slope at a point on a circle
A curve you cannot solve for y
Implicit second derivatives
What inverse means, geometrically
The inverse derivative formula
- If you want (f⁻¹)'(b), find the input a such that f(a) = b, then it's 1/f'(a).
- Geometric: reciprocal slopes of perpendicular lines.
Find (f⁻¹)'(b) without finding f⁻¹
x1/n via the inverse theorem
The six inverse trig functions
arcsin x
"the angle whose sine is x." Range [−π/2, π/2].arccos x
Range [0, π].arctan x
Range (−π/2, π/2). Defined for all real x.arccot x
Range (0, π).arcsec x
Range [0, π], y ≠ π/2. Defined for |x| ≥ 1.arccsc x
Range [−π/2, π/2], y ≠ 0. Defined for |x| ≥ 1.Six inverse trig derivatives · memorize four
| Function | Derivative | Domain restriction |
|---|---|---|
| arcsin x | 1 / √(1 − x²) | −1 < x < 1 |
| arccos x | −1 / √(1 − x²) | −1 < x < 1 |
| arctan x | 1 / (1 + x²) | all real x |
| arccot x | −1 / (1 + x²) | all real x |
| arcsec x | 1 / (|x| √(x² − 1)) | |x| > 1 |
| arccsc x | −1 / (|x| √(x² − 1)) | |x| > 1 |
Derive d/dx [arcsin x]
Chain rule with inverse trig
Why inverse trig matters in integration
Where the chain rule meets real life
Chain rule practice run
Implicit diff practice run
Inverse function derivative practice
Quick check — chain rule
Click an option to check.
AP MCQ · implicit differentiation
Click to check.
AP FRQ · inverse function
AP FRQ · implicit second derivative
Top 5 mistakes in Unit 3
① Dropping the inner derivative
"d/dx [sin(x²)] = cos(x²)" — missing the ·2x. Always multiply by the derivative of the inside.
② Forgetting to apply the chain rule to y when implicit-diffing
If your equation has y, y², sin(y), etc., you must bring dy/dx along with the chain rule. Example: differentiating y² gives 2y · dy/dx, not just 2y.
③ Wrong sign on arcsin' / arctan'
arcsin' is positive. arccos' is NEGATIVE. arctan' is positive. arccot' is negative. Don't memorize just two of them.
④ Using x = a vs x = (f⁻¹)(a) carelessly
The inverse function theorem mixes the two arguments. Be sure: (f⁻¹)'(b) = 1 / f'(a) where a = f⁻¹(b) (i.e., f(a) = b).
⑤ Differentiating arctan as 1/x instead of 1/(1+x²)
1/x is d/dx [ln x]. 1/(1+x²) is d/dx [arctan x]. Different formulas — different functions.
Chain rule, implicit diff, and inverses in the wild
- Arctan in machine learning: The activation function arctan(x) is differentiable everywhere (unlike ReLU), so its gradient never vanishes. d/dx arctan(x) = 1/(1+x²) appears in millions of neural-network updates per second.
- Implicit curves in design: Bézier curves (used in every font, logo, UI animation) are defined implicitly. Designers use calculus-of-implicit-curves to control lighting, shadows, and 3D motion.
- Inverse function theorem in physics: The Jacobian of an inverse transformation (e.g., coordinate changes in fluid dynamics) gives you the inverse derivative framework. Critical for general relativity.
- Drug concentration: If concentration C(t) depends on time, and you want to know how dose rate affects peak concentration, the chain rule through the inverse function lets you solve for t as a function of C.
- Stock options pricing: The Black-Scholes formula uses arctan-like terms and implicit relationships between strike price, time, and volatility.
Best YouTube for this unit
Unit 3 recap
- Chain rule: (f∘g)' = f'(g(x)) · g'(x). When in doubt, peel layers.
- Implicit diff: Differentiate both sides w.r.t. x; bring dy/dx along via chain rule; solve.
- Inverse function theorem: (f⁻¹)'(b) = 1 / f'(a) where b = f(a).
- Inverse trig derivatives: arcsin' = 1/√(1−x²), arccos' = −1/√(1−x²), arctan' = 1/(1+x²), arccot' = −1/(1+x²).
- Composite inverse derivative = 1 / (composite of original derivative evaluated at the inverse).
Unit 3 · everything covered ✓
Mapped 1:1 against the 2024-25 College Board CED (across Topic 3 sub-skills).
- 3.1 / FUN-3.C Composite functions & chain rule (formal statement).
- 3.1 / FUN-3.C Chain rule in tables (3+ layer chain).
- 3.1 / FUN-3.C Combining chain rule with product, quotient, sum rules.
- 3.2 / FUN-3.D Implicit differentiation — first derivative.
- 3.2 / FUN-3.D Implicit differentiation — second derivatives.
- 3.3 / FUN-3.E Inverse function theorem.
- 3.3 / FUN-3.E Derivatives of inverse functions of xn.
- 3.4 / FUN-3.F Derivatives of arcsin, arccos, arctan, arccot, arcsec, arccsc.
- 3.5 / FUN-3.* Inverse trig + chain rule combined.
- 3 bonus Related rates preview (full lesson in Unit 4).
- Worked examples · 4 unlimited + 3 quiz · 5 practice problems embedded.
- Common errors · 5 pitfalls explicitly called out.
- 3 curated YouTube lessons (chain / implicit / inverse trig).
- Real-world connections · 5 applications across ML, design, physics, medicine, finance.