Self-Assessment Quiz: Cardinality and Infinity
Twenty questions to check your understanding. Answer before opening the key. Aim for 16+. Throughout, $\mathbb{N} = \{0, 1, 2, \dots\}$ includes 0.
Question 1
To prove two sets $A$ and $B$ have the same cardinality, you must exhibit:
A) a count of the elements of each B) an injection $A \to B$ C) a bijection $A \to B$ D) a surjection $A \to B$
Question 2
A set is countably infinite exactly when:
A) it is a subset of $\mathbb{N}$ B) it is in bijection with $\mathbb{N}$ C) it is finite D) its elements can be listed in increasing order
Question 3
Which set is uncountable?
A) $\mathbb{Z}$ B) $\mathbb{Q}$ C) $\mathbb{N} \times \mathbb{N}$ D) $\mathbb{R}$
Question 4
The bijection $f(n) = 2n$ from $\mathbb{N}$ to the even naturals proves:
A) the evens are a proper subset of $\mathbb{N}$ B) a set can be equinumerous with a proper subset of itself C) the evens are uncountable D) $\mathbb{N}$ is finite
Question 5
"A set is infinite iff it is equinumerous with a proper subset of itself." This is:
A) false — no set has that property B) Dedekind's definition of an infinite set C) true only for $\mathbb{R}$ D) the definition of countable
Question 6
In §10.2, the anti-diagonal sweep of the grid is used to prove:
A) $\mathbb{R}$ is uncountable B) $\mathbb{N} \times \mathbb{N}$ (and hence $\mathbb{Q}$) is countable C) the power set is larger than the set D) the halting problem is undecidable
Question 7
$\mathbb{Q}$ is dense (between any two rationals lies another). Why does this not make $\mathbb{Q}$ uncountable?
A) density implies uncountability, so the premise is wrong B) density is about order, not size; a countable set can be dense in an uncountable one C) $\mathbb{Q}$ is actually finite D) only $\mathbb{R}$ is dense
Question 8
In Cantor's diagonal proof, the constructed number $x$ cannot equal $r_n$ because:
A) $x$ is irrational and every $r_n$ is rational B) $x$ differs from $r_n$ in its $n$-th decimal digit C) $x$ is larger than every $r_n$ D) $x$ is not in $[0, 1)$
Question 9
Why does the diagonal proof forbid decimal expansions ending in an infinite tail of 9s?
A) to keep $x$ in $[0,1)$ B) to make each real's decimal expansion unique, so "different digit" means "different number" C) because 9 is not a valid digit D) it is an arbitrary convention with no effect
Question 10
The set of all programs in a fixed language is countable because:
A) there are finitely many programs B) each program is a finite string over a finite alphabet, and such strings can be listed by length then alphabetically C) programs are real numbers D) most programs don't compile
Question 11
The set of all decision problems ($f\colon \mathbb{N} \to \{0,1\}$) is uncountable. The proof is:
A) the anti-diagonal sweep B) the pigeonhole principle C) a diagonal bit-flip: $g(n) = 1 - f_n(n)$ escapes any proposed list D) induction on $n$
Question 12
"There are more problems than programs" forces which conclusion?
A) every problem has a program B) some well-defined problems are computed by no program C) programs are uncountable D) the rationals are uncountable
Question 13
The set of functions $\mathbb{N} \to \{0,1\}$ corresponds exactly to which Chapter 8 object?
A) the Cartesian product $\mathbb{N} \times \mathbb{N}$ B) the power set $\mathcal{P}(\mathbb{N})$ C) the empty set D) the set of finite subsets of $\mathbb{N}$
Question 14
$\aleph_0$ denotes:
A) the cardinality of $\mathbb{R}$ B) the cardinality of any countably infinite set — the smallest infinite cardinal C) the largest infinity D) zero
Question 15
Cantor's theorem states that for every set $S$:
A) $\lvert S\rvert = \lvert \mathcal{P}(S)\rvert$ B) $\lvert S\rvert < \lvert \mathcal{P}(S)\rvert$ C) $\lvert S\rvert > \lvert \mathcal{P}(S)\rvert$ D) $S$ is countable
Question 16 (True/False, justify)
True or false: If $x$ is the number produced by Cantor's diagonal argument on a list $r_0, r_1, \dots$, then inserting $x$ at the front of the list makes the list complete. Justify in one sentence.
Question 17 (True/False, justify)
True or false: $\aleph_0 + \aleph_0 = \aleph_0$. Name a set that witnesses your answer.
Question 18 (True/False, justify)
True or false: Because IEEE-754 double represents only about $2^{64}$ values while $\mathbb{R}$ is
uncountable, almost every real number is not exactly representable as a double. Justify in one
sentence.
Question 19 (Short answer)
In one sentence each, distinguish the two "diagonal" moves of this chapter: sweeping along anti-diagonals versus walking down the diagonal. What does each prove?
Question 20 (Short answer)
State, in one sentence, why the existence of uncomputable problems is a counting fact that holds before we ever define precisely what a "program" or "computation" is.
Answer Key
| Q | Ans | Note |
|---|---|---|
| 1 | C | Same cardinality = a bijection exists. |
| 2 | B | Countably infinite = in bijection with $\mathbb{N}$. |
| 3 | D | $\mathbb{R}$ is uncountable; $\mathbb{Z}, \mathbb{Q}, \mathbb{N}\times\mathbb{N}$ are countable. |
| 4 | B | The signature of an infinite set (Dedekind). |
| 5 | B | Dedekind's definition of "infinite." |
| 6 | B | The sweep gathers every grid cell into one list ⇒ countability. |
| 7 | B | Density concerns order; a countable set can be dense in an uncountable one. |
| 8 | B | $x$ was built so $x_n \ne d_{nn}$; differing in any digit ⇒ unequal. |
| 9 | B | Uniqueness of expansion makes "different digits" genuinely mean "different number." |
| 10 | B | Finite strings over a finite alphabet list by length then alphabetically. |
| 11 | C | The diagonal bit-flip escapes every list ⇒ uncountable. |
| 12 | B | Countable $<$ uncountable ⇒ no surjection programs $\to$ problems. |
| 13 | B | A function $\mathbb{N}\to\{0,1\}$ is the indicator of a subset ⇒ $\mathcal{P}(\mathbb{N})$. |
| 14 | B | The smallest infinite cardinal. |
| 15 | B | The power set is always strictly larger. |
| 16 | False | Re-running the construction on the patched list yields a new escapee; every list misses something. |
| 17 | True | Witnessed by $\mathbb{Z}$ (evens + odds), or $\mathbb{N} = E \cup (\mathbb{N}\setminus E)$. |
| 18 | True | A finite (hence countable) value set cannot cover an uncountable line; the unrepresentable reals are "almost all." |
| 19 | — | Along: gather everything into one list ⇒ countability ($\mathbb{Q}$, $\mathbb{N}\times\mathbb{N}$). Down: construct something on no list ⇒ uncountability ($\mathbb{R}$, $\mathcal{P}(\mathbb{N})$). |
| 20 | — | Programs are countable and problems are uncountable as sets; the strict inequality $\aleph_0 < \lvert\text{problems}\rvert$ alone forces a problem with no program, independent of any model of computation. |
Topics to review by question
| Questions | Topic |
|---|---|
| 1, 2, 14 | Definitions: same size, countable, $\aleph_0$ (§10.1–10.2, 10.5) |
| 3, 6, 7 | Which sets are countable; the grid sweep (§10.2) |
| 4, 5 | Part-equals-whole; Dedekind-infinite (§10.1) |
| 8, 9, 16 | Cantor diagonalization mechanics and the "just add $x$" pitfall (§10.3) |
| 10, 11, 12, 20 | More problems than programs (§10.4) |
| 13, 15 | Functions $\mathbb{N}\to\{0,1\}$, power sets, Cantor's theorem (§10.4–10.5) |
| 17 | Cardinal arithmetic / Hilbert's Hotel (§10.5) |
| 18 | Why a programmer cares; floating point (§10.6) |
| 19 | The two diagonal moves, kept distinct (Summary) |