Inverses of Exponential Functions - AP Precalculus

SECTION 1

Historical Context & Motivation

Long before the formal language of functions and inverses existed, mathematicians and astronomers grappled with a practical problem: given a table of powers, how could one work backward to find the exponent that produced a known result? The need to reverse exponential relationships drove the invention of one of mathematics' most important tools—the logarithm. Exponential functions model phenomena such as compound interest, radioactive decay, and population growth, yet answering the question "how long until a quantity reaches a particular level?" requires an inverse operation. This interplay between exponential and logarithmic functions forms the conceptual backbone of the AP Precalculus unit on exponential and logarithmic functions.

1614 • Napier Publishes Mirifici Logarithmorum

John Napier introduced logarithms as a computational device to simplify multiplication and division into addition and subtraction, effectively constructing the first systematic inverse of an exponential relationship.

1624 • Briggs' Common Logarithm Tables

Henry Briggs refined Napier's work by establishing base-10 logarithm tables, making the inverse of 10ˣ accessible to navigators, engineers, and scientists across Europe.

1748 • Euler's Introductio in Analysin Infinitorum

Leonhard Euler formalized the exponential function eˣ and its inverse, the natural logarithm ln x, embedding them as core functions in mathematical analysis and connecting them through the concept of inverse functions.

1821 • Cauchy's Rigorous Function Theory

Augustin-Louis Cauchy placed inverse functions on firm theoretical ground, formalizing the conditions under which a function's inverse exists—particularly the requirement that the original function be one-to-one.

The central question this lesson addresses is both simple to state and rich in consequences: given an exponential function f(x) = bˣ, what function undoes it—that is, what function takes an output of f and recovers the original input? Answering this question leads directly to the logarithmic function, and understanding why and how logarithms serve as inverses of exponentials is essential for solving equations, interpreting models, and reasoning about rates of change throughout this course and beyond.

SECTION 2

Core Principles & Definitions

Before deriving the inverse of an exponential function, it is essential to recall what it means for two functions to be inverses of one another and why exponential functions satisfy the conditions that guarantee an inverse exists. An inverse function f⁻¹ reverses the action of f: if f(a) = b, then f⁻¹(b) = a. Equivalently, (f⁻¹ ∘ f)(x) = x for every x in the domain of f, and (f ∘ f⁻¹)(x) = x for every x in the domain of f⁻¹. For an inverse to exist as a function, the original function must be one-to-one (injective): distinct inputs must produce distinct outputs. Exponential functions of the form f(x) = bˣ with b > 0 and b ≠ 1 are strictly monotonic—either always increasing (b > 1) or always decreasing (0 < b < 1)—and therefore pass the horizontal line test, guaranteeing that an inverse function exists.

One-to-One Property

An exponential function f(x) = bˣ (b > 0, b ≠ 1) is strictly monotonic: if b > 1 it is always increasing, and if 0 < b < 1 it is always decreasing. No two distinct inputs share the same output, so f passes the horizontal line test and its inverse is a function.

Domain–Range Swap

The domain of f(x) = bˣ is all real numbers (−∞, ∞) and its range is (0, ∞). In the inverse, these exchange: the domain of the logarithmic function

\log_b

(x) is (0, ∞) and its range is (−∞, ∞).

Defining the Logarithm

The inverse of f(x) = bˣ is defined as f⁻¹(x) =

\log_b

(x). By definition, y =

\log_b

(x) if and only if bʸ = x. The logarithm answers the question: to what power must the base b be raised to produce x?

Inverse Composition Identities

The exponential and logarithmic functions undo each other:

b^{log_b(x)}

= x for all x > 0, and

\log_b

(bˣ) = x for all real x. These identities are the operational definition of inverse functions applied to this specific pair.

Graphical Reflection

The graphs of f(x) = bˣ and f⁻¹(x) =

\log_b

(x) are reflections of each other across the line y = x. Each point (a, b) on one graph corresponds to the point (b, a) on the other.

✦

KEY TAKEAWAY

Think of an exponential function as a locking mechanism: given a base b, the function bˣ "locks" any real number x into a positive output. The logarithm

\log_b

is the unique key that unlocks that output, recovering the original exponent. Just as decryption reverses encryption in a secure communication protocol, the logarithm reverses exponentiation—and the guarantee that this key is unique comes from the one-to-one nature of exponential functions.

SECTION 3

Visual Explanation — Graphs of Inverse Pairs

The defining geometric relationship between an exponential function and its logarithmic inverse is reflection across the line y = x. The diagram below plots f(x) = 2ˣ (in cyan) alongside its inverse f⁻¹(x) = log₂(x) (in pink), with the mirror line y = x shown as a dashed reference. Notice how the exponential's horizontal asymptote at y = 0 becomes a vertical asymptote at x = 0 for the logarithm, and how key anchor points swap their coordinates.

The cyan curve represents f(x) = 2ˣ with its horizontal asymptote at y = 0. The pink curve represents f⁻¹(x) = log₂(x) with its vertical asymptote at x = 0. Each labeled point on one curve has a reflected counterpart—with swapped coordinates—on the other curve. The dashed line y = x serves as the axis of reflection.

Observe several important features in the diagram. The point (0, 1) on the exponential curve reflects to (1, 0) on the logarithmic curve—this is why every logarithmic function passes through (1, 0). The horizontal asymptote of the exponential (y = 0) transforms into the vertical asymptote of the logarithm (x = 0), a consequence of the domain–range swap. As x → −∞ the exponential approaches 0 from above, so correspondingly as x → 0⁺ the logarithm plunges toward −∞. These graphical features are not coincidental; they are guaranteed by the algebraic structure of inverse functions.

SECTION 4

Mathematical Framework

Deriving the inverse of an exponential function algebraically follows the standard procedure: write y = f(x), swap x and y, and solve for y. This process makes explicit why the logarithm is the natural result of inverting exponentiation.

EXPONENTIAL FUNCTION

f(x) = bˣ where b > 0 and b ≠ 1

b is the base of the exponential. The domain is (−∞, ∞) and the range is (0, ∞).

INVERSE DERIVATION

y = bˣ → x = bʸ → y = log_b(x)

Step 1: Start with y = bˣ. Step 2: Swap x and y to get x = bʸ. Step 3: Solve for y by applying log base b to both sides, yielding y = $\log_b$ (x). Therefore f⁻¹(x) = $\log_b$ (x).

INVERSE COMPOSITION IDENTITIES

b^(log_b(x)) = x for x > 0 and log_b(bˣ) = x for all real x

These identities confirm that bˣ and $\log_b$ (x) are true inverses. The first says applying the exponential to a logarithm recovers the original input; the second says applying the logarithm to an exponential output recovers the exponent.

GENERAL EXPONENTIAL INVERSE

If g(x) = a · b^(cx + d) + k, then g⁻¹(x) = [log_b((x − k)/a) − d] / c

For transformed exponentials, undoing the operations in reverse order gives the inverse. Subtract k, divide by a, take log base b, subtract d, and divide by c. This inverse exists when a ≠ 0, b > 0, b ≠ 1, c ≠ 0, and (x − k)/a > 0.

The derivation of the general inverse for a transformed exponential illustrates an important principle: to find the inverse, you undo each transformation in reverse order. If the exponential function first multiplies the input by c, then adds d, then raises b to that power, then multiplies by a, and finally adds k, the inverse must subtract k first, then divide by a, then apply the logarithm, then subtract d, and finally divide by c. This "peeling off layers" approach is especially useful on the AP Precalculus exam, where transformed exponential functions frequently appear in both multiple-choice and free-response contexts.

⚠

Domain Restriction Alert

When finding the inverse of g(x) = a ·

b^{cx + d}

+ k, the vertical shift k becomes a critical domain restriction for the inverse. Since bˣ > 0 for all x, if a > 0 then g(x) > k, so the domain of g⁻¹ is (k, ∞). If a < 0, then g(x) < k, and the domain of g⁻¹ is (−∞, k). Failing to state this restriction is a common point-loss on FRQs.

SECTION 5

Properties of Logarithmic Functions as Inverses

Because logarithmic functions are defined as inverses of exponential functions, every property of exponentials has a reflected counterpart in logarithms. The table below systematically catalogs these parallel properties, reinforcing the inverse relationship. Understanding these correspondences is essential for analyzing the behavior of logarithmic models in context.

Property	Exponential f(x) = bˣ (b > 1)	Logarithmic f⁻¹(x) = $\log_b$ (x) (b > 1)
Domain	(−∞, ∞)	(0, ∞)
Range	(0, ∞)	(−∞, ∞)
Key Point	(0, 1)	(1, 0)
Asymptote	Horizontal: y = 0	Vertical: x = 0
Monotonicity	Strictly increasing	Strictly increasing
End Behavior	As x → −∞, f(x) → 0⁺; as x → ∞, f(x) → ∞	As x → 0⁺, f⁻¹(x) → −∞; as x → ∞, f⁻¹(x) → ∞
Concavity	Concave up (increasing at an increasing rate)	Concave down (increasing at a decreasing rate)

Parallel properties of exponential and logarithmic functions

Side-by-side comparison of f(x) = 2ˣ (left, cyan border) and f⁻¹(x) = log₂(x) (right, pink border). The domain–range badges beneath each graph show the swap: the exponential's range (0, ∞) becomes the logarithm's domain, and vice versa. The horizontal asymptote transforms into a vertical asymptote, and the anchor point coordinates reflect.

The concavity correspondence deserves particular attention because it connects to rates of change. The exponential f(x) = 2ˣ is concave up, meaning its outputs increase at an increasing rate—each unit increase in x doubles the output. The inverse, log₂(x), is concave down: its outputs increase at a decreasing rate. Doubling the input to the logarithm only adds 1 to the output. This diminishing-returns behavior is why logarithmic scales (such as the Richter scale or decibel scale) are used to compress enormous ranges of values into a manageable scale.

SECTION 6

Worked Example

Let us find the inverse of a transformed exponential function, a task that integrates all the principles developed so far. Consider g(x) = 3 · $2^{x − 1}$ + 5. We will derive g⁻¹(x) algebraically, state its domain and range, and verify the result using the composition identities.

Finding the Inverse of g(x) = 3 · $2^{x − 1}$ + 5

1
Step 1 — Set y = g(x) and swap x and y

Begin by writing y = 3 · $2^{x − 1}$ + 5. To find the inverse, swap x and y: x = 3 · $2^{y − 1}$ + 5. Our task is now to solve this equation for y.
2
Step 2 — Isolate the exponential expression

Subtract 5 from both sides: x − 5 = 3 · $2^{y − 1}$ . Then divide both sides by 3: (x − 5)/3 = $2^{y − 1}$ . Note that for this expression to be valid, we need (x − 5)/3 > 0, which gives x > 5.

(x − 5)/3 = $2^{y − 1}$ with x > 5
3
Step 3 — Apply the logarithm to both sides

Take log₂ of both sides: log₂((x − 5)/3) = y − 1. This step uses the definition that log₂ is the inverse of 2ˣ, so log₂( $2^{y − 1}$ ) = y − 1.

log₂((x − 5)/3) = y − 1
4
Step 4 — Solve for y

Add 1 to both sides: y = log₂((x − 5)/3) + 1. Therefore g⁻¹(x) = log₂((x − 5)/3) + 1.

g⁻¹(x) = log₂((x − 5)/3) + 1
5
Step 5 — State the domain and range of g⁻¹

The original function g has domain (−∞, ∞) and range (5, ∞), since 3 · $2^{x − 1}$ > 0 means g(x) > 5 for all x. By the domain–range swap, g⁻¹ has domain (5, ∞) and range (−∞, ∞). The vertical asymptote of g⁻¹ is the line x = 5.

Domain of g⁻¹: (5, ∞) Range of g⁻¹: (−∞, ∞)
6
Step 6 — Verify via composition

Check: g(g⁻¹(x)) = 3 · 2^((log₂((x−5)/3) + 1) − 1) + 5 = 3 · 2^(log₂((x−5)/3)) + 5 = 3 · (x−5)/3 + 5 = (x−5) + 5 = x. ✓ The composition returns x, confirming the inverse is correct.

g(g⁻¹(x)) = x ✓

SECTION 7

Common Exponential–Logarithm Pairs & Pitfalls

While the inverse relationship holds for any valid base b, three exponential–logarithm pairs appear most frequently in AP Precalculus and deserve special familiarity. The table below compares these pairs and highlights their typical contexts. Understanding the notational conventions for each is essential for avoiding errors on the exam.

Exponential Function	Inverse (Logarithmic)	Notation	Common Context
f(x) = 10ˣ	f⁻¹(x) = log₁₀(x)	log x (common log)	pH scale, Richter scale, decibels
f(x) = eˣ	f⁻¹(x) = ln(x)	ln x (natural log)	Continuous growth/decay, calculus, half-life
f(x) = 2ˣ	f⁻¹(x) = log₂(x)	log₂ x (binary log)	Computer science, doubling-time problems

Three most common exponential–logarithm inverse pairs

Common Pitfalls

Forgetting domain restrictions. The argument of any logarithm must be strictly positive. When finding the inverse of a transformed exponential g(x) = a · bˣ + k, the inverse's domain is restricted by the vertical shift k and the sign of a.
Confusing log notation. On the AP exam, "log x" without a subscript means log₁₀(x), while "ln x" means logₑ(x). Some textbooks use "log" for the natural logarithm—always check context.
Misapplying the inverse to non-one-to-one functions. The function f(x) = b²ˣ is still one-to-one and has an inverse. However, if a function like h(x) = |bˣ − 4| is given, it may not be one-to-one, and its inverse would not exist without a domain restriction.
Algebraic order of operations when unwinding transformations. Remember that the inverse undoes operations in reverse order. A common error is to apply the logarithm before isolating the exponential term.

✦

KEY TAKEAWAY

The logarithm is not a separate, unrelated function—it is structurally defined as the inverse of the exponential. Every property of logarithms (product rule, quotient rule, power rule) can be derived directly from exponential properties by applying the inverse relationship. Mastering this connection means you only need to deeply learn one family of functions to understand both.

SECTION 8

Connections to Advanced Topics

The inverse relationship between exponential and logarithmic functions serves as a foundation for numerous concepts that extend beyond AP Precalculus. Recognizing these forward connections helps you appreciate why this topic receives such emphasis and how the skills developed here transfer to higher mathematics and applied sciences.

AP Precalculus Concept	Advanced Extension	How the Inverse Is Used
Solving bˣ = c for x	Calculus: solving differential equations	The natural logarithm isolates exponents in solutions to dy/dx = ky, yielding y = $Ce^{kt}$ and t = ln(y/C)/k.
Graph of $\log_b$ (x) as reflection of bˣ	Calculus: derivative of ln x = 1/x	The derivative of the inverse function is the reciprocal of the original's derivative evaluated at the reflected point: d/dx[ln x] = 1/( $e^{ln x}$ ) = 1/x.
Semi-log regression	Statistics: log-transformations	Applying the logarithm (inverse) to exponentially distributed data linearizes it, enabling linear regression analysis on the log-transformed scale.
Change of base formula	Information theory: entropy (log₂)	Converting between logarithmic bases via the inverse relationship allows measuring information in bits (base 2), nats (base e), or bans (base 10).

Forward connections from AP Precalculus to advanced coursework

Perhaps the most elegant advanced connection is the way the inverse function theorem in calculus uses the relationship between eˣ and ln x to derive the derivative of the natural logarithm. Because (d/dx)(eˣ) = eˣ, the inverse function theorem gives (d/dx)(ln x) = 1/eˡⁿˣ = 1/x. This clean result—one of the most important in all of calculus—flows directly from the inverse relationship you are mastering now. In AP Precalculus, you build the foundational understanding of what the inverse does; calculus later asks how fast the inverse changes.

SECTION 9

Practice Problems

Question 1: CONCEPTUAL

Which of the following best explains why the function f(x) = 5ˣ has an inverse that is also a function?

A. Because f(x) = 5ˣ is defined for all real numbers x.
B. Because f(x) = 5ˣ is strictly increasing, meaning it is one-to-one.
C. Because f(x) = 5ˣ is continuous on its domain.
D. Because the range of f(x) = 5ˣ is (0, ∞), which does not include negative numbers.
E. Because 5ˣ passes the vertical line test.

Show answer (B)

The correct answer is (B). For an inverse to exist as a function, the original function must be one-to-one (injective), which means it must pass the horizontal line test. Since f(x) = 5ˣ is strictly increasing (because b = 5 > 1), every horizontal line intersects the graph at most once, confirming the one-to-one property. (A) is incorrect because being defined for all real numbers describes the domain, not injectivity—many non-one-to-one functions have domain (−∞, ∞). (C) is incorrect because continuity alone does not guarantee an inverse; for example, f(x) = x² is continuous but not one-to-one. (D) is incorrect because having a range of (0, ∞) describes the output values but does not address whether different inputs can produce the same output. (E) is incorrect because the vertical line test determines whether a relation is a function, not whether it has an inverse; all functions pass the vertical line test.

Question 2: BASIC CALCULATION

What is the inverse of f(x) = 4ˣ − 7?

A. f⁻¹(x) = log₄(x + 7)
B. f⁻¹(x) = log₄(x − 7)
C. f⁻¹(x) = log₄(x) + 7
D. f⁻¹(x) = $4^{x + 7}$
E. f⁻¹(x) = log₇(x + 4)

Show answer (A)

The correct answer is (A). To find the inverse, write y = 4ˣ − 7, swap x and y to get x = 4ʸ − 7, then solve: x + 7 = 4ʸ, so y = log₄(x + 7). Therefore f⁻¹(x) = log₄(x + 7) with domain (−7, ∞). (B) subtracts 7 instead of adding it, failing to correctly undo the subtraction of 7 in the original function. (C) applies the logarithm first and then adds 7, but the addition should occur inside the logarithm argument (you must isolate the exponential before taking the log). (D) gives another exponential rather than a logarithmic inverse. (E) uses incorrect base and argument.

Question 3: INTERMEDIATE

The function h(x) = −2 ·

3^{x + 4}

+ 10 is defined for all real numbers. What are the domain and vertical asymptote of h⁻¹(x)?

A. Domain: (10, ∞); vertical asymptote: x = 10
B. Domain: (−∞, 10); vertical asymptote: x = 10
C. Domain: (−∞, ∞); vertical asymptote: x = −4
D. Domain: (−∞, −10); vertical asymptote: x = −10
E. Domain: (0, ∞); vertical asymptote: x = 0

Show answer (B)

The correct answer is (B). Since the leading coefficient a = −2 is negative and

3^{x+4}

> 0 for all real x, we have −2 ·

3^{x+4}

< 0, so h(x) = −2 ·

3^{x+4}

+ 10 < 10 for all x. Thus the range of h is (−∞, 10). By the domain–range swap, the domain of h⁻¹ is (−∞, 10), and the horizontal asymptote y = 10 of h becomes the vertical asymptote x = 10 of h⁻¹. (A) incorrectly assumes the range is (10, ∞), which would be the case if the coefficient were positive. (C) is wrong because the inverse of an exponential function cannot have domain (−∞, ∞)—the logarithmic argument must be restricted. (D) uses −10 rather than 10. (E) ignores the transformations entirely and defaults to the untransformed case.

Question 4: APPLIED

4 pts

A biologist models the population of a bacterial colony as P(t) = 500 ·

2^{t/3}

, where P is the number of bacteria and t is time in hours after inoculation. (a) Find the inverse function P⁻¹(x) and interpret its meaning in context. (b) State the domain and range of P⁻¹ and explain their significance. (c) Use P⁻¹ to determine how many hours after inoculation the population reaches 16,000 bacteria. (d) Verify your answer to part (c) by substituting back into the original function.

Show answer

Rubric (4 points): Point 1 — Part (a): Finding and interpreting the inverse. Write x = 500 ·

2^{y/3}

. Divide: x/500 =

2^{y/3}

. Take log₂: log₂(x/500) = y/3. Multiply by 3: y = 3 · log₂(x/500). So P⁻¹(x) = 3 · log₂(x/500). Interpretation: P⁻¹(x) gives the number of hours after inoculation at which the population first reaches x bacteria. Point 2 — Part (b): Domain and range with contextual significance. Since P(t) has domain [0, ∞) (time cannot be negative) and range [500, ∞) (population starts at 500 at t = 0), the inverse P⁻¹ has domain [500, ∞) and range [0, ∞). In context, the domain restriction x ≥ 500 means it only makes sense to ask "when does the population reach x" for values at or above the initial population of 500. Point 3 — Part (c): Correct computation. P⁻¹(16000) = 3 · log₂(16000/500) = 3 · log₂(32) = 3 · 5 = 15 hours. Point 4 — Part (d): Verification. P(15) = 500 ·

2^{15/3}

= 500 · 2⁵ = 500 · 32 = 16000. ✓ This confirms the answer.

Question 5: CRITICAL THINKING

3 pts

Consider the function f(x) =

b^{g(x)}

where b > 1 and g is a strictly increasing linear function g(x) = mx + d with m > 0. (a) Prove that f is one-to-one. (b) Derive f⁻¹(x) in terms of b, m, and d. (c) Explain how the parameters m and d each affect the graph of f⁻¹ relative to the graph of

\log_b

(x).

Show answer

Rubric (3 points): Point 1 — Part (a): Proof of one-to-one property. Suppose f(x₁) = f(x₂), i.e.,

b^{mx₁ + d}

b^{mx₂ + d}

. Since the exponential function with base b > 1 is strictly increasing (and therefore one-to-one), we have mx₁ + d = mx₂ + d, which gives mx₁ = mx₂, and since m > 0, x₁ = x₂. Thus f is one-to-one. (Alternatively: f is the composition of a strictly increasing linear function g and a strictly increasing exponential function bˣ, and the composition of strictly increasing functions is strictly increasing, hence one-to-one.) Point 2 — Part (b): Correct derivation of f⁻¹. Set y =

b^{mx + d}

. Swap: x =

b^{my + d}

. Take

\log_b

\log_b

(x) = my + d. Solve for y: y = (

\log_b

(x) − d)/m. Therefore f⁻¹(x) = (1/m) ·

\log_b

(x) − d/m, with domain (0, ∞). Point 3 — Part (c): Analysis of parameter effects. The parameter m produces a vertical compression of

\log_b

(x) by the factor 1/m: larger m values compress the graph vertically, making the inverse grow more slowly. The parameter d produces a vertical shift of −d/m: if d > 0, the graph of f⁻¹ shifts downward by d/m units relative to

\log_b

(x); if d < 0, it shifts upward. Together, these correspond to the fact that in the original function, m controls the horizontal stretch of the exponential (steeper growth for larger m) and d controls the horizontal shift.

SUMMARY

Summary & Review

Summary

The inverse of an exponential function f(x) = bˣ is the logarithmic function f⁻¹(x) = $\log_b$ (x), defined by the equivalence y = $\log_b$ (x) ⟺ bʸ = x. This inverse exists because exponential functions are strictly monotonic and therefore one-to-one. The domain and range swap between a function and its inverse means that the domain of bˣ—all real numbers—becomes the range of $\log_b$ (x), and the range of bˣ—the positive reals—becomes the domain of $\log_b$ (x). Graphically, the two curves are reflections across the line y = x, with the horizontal asymptote of the exponential mapping to the vertical asymptote of the logarithm.

For transformed exponentials such as g(x) = a · $b^{cx + d}$ + k, the inverse is found by undoing each transformation in reverse order: subtract k, divide by a, take $\log_b$ , subtract d, and divide by c. The composition identities $b^{log_b(x)}$ = x and $\log_b$ (bˣ) = x provide a powerful verification tool. Mastery of these ideas prepares you not only for solving exponential equations on the AP exam but also for the deeper study of logarithmic properties, semi-log models, and the eventual calculus of exponential and logarithmic functions.