Week 11 - Mathematical Induction A

6.101 Introduction to proofs

  • Proof
    • A proof is a valid argument, that is used to prove the truth of a statement.
    • To build a proof, we need to use things previously introduced:
      • variables and predicates.
      • quantifiers.
      • laws of logic.
      • rules of inference.
    • Some terminology used:
      • Theorem
        • formal statement that can be shown to be true.
      • Axiom
        • a statement we assume to be true to serve as a premise for further arguments.
      • Lemma
        • a proven statement used as a step to a larger result rather than as a statement of interest by itself.
  • Formalising a theorem
    • Consider statement S: "There exists a real number between any two not equal real numbers"
    • S can be formalised as: x,yR\forall x, y \in \mathbb{R} if x<yx < y then zR\exists z \in \mathbb{R} where x<z<yx < z < y.
    • S is an example of a theorem.
  • Direct proof
    • A direct proof is based on showing that a conditional statement: pqp \rightarrow q is true.
    • We start by assuming that p is true and then use: Axioms, definitions and Theorems, together with rules of inference to show that q must also be true.
    • Example
      • "There exists a real number between any two not equal real numbers"
      • Proof:
        • Let x,yx, y be arbitrary elements in R\mathbb{R}
        • Let's suppose x<yx < y
        • Let z=(x+y)/2z = (x + y) / 2
        • zRz \in \mathbb{R}, satisfying x<z<yx < z < y
      • Therefore, using the universal generalised rule, we can conclude that: x,yR\forall x, y \in \mathbb{R} if x<yx < y then zR\exists z \in \mathbb{R} where x<z<yx < z < y.
  • Proof by Contrapositive
    • A proof by contrapositive is based on the fact that proving the conditional statement pqp \rightarrow q is equivalent to providing its contrapositive ¬q¬p\neg q \rightarrow \neg p
      • We start by assuming that ¬q\neg q is true and then use axioms, definitions and theorems, together with rules of inference, to show that ¬p\neg p must be true.
    • Example
      • Proof of the theorem:
        • If n2n^2 is even then n is even
        • Direct proof:
          • Let nZn \in Z. If n2n^2 is even then kZ\exists k \in Z, n2=2kn^2 = 2k
          • Then kZ\exists k \in Z, n=±2kn = \pm \sqrt{2k}
            • From this equation it doesn't seem intuitive to prove that n is even.
        • Proof by contraposition:
          • Let's supposed n is odd.
          • Then kZ\exists k \in Z, n=2k+1n = 2k + 1
          • Then kZ\exists k \in Z, n2=(2k+1)2=2(2k2+2k)+1n^2 = (2k + 1)^2 = 2(2k^2 + 2k) + 1
          • Then n2n^2 is also odd.
          • We have succeeded in providing the contrapositive: if n is odd then n2n^2 is odd.
  • Proof by contradiction
    • A proof by contradiction is based on assuming that the statement we want to prove is false, and then showing that this assumption leads to a false proposition.
    • We start by assuming that ¬p\neg p is true and then use: Axioms, definitions and Theorems, together with rules of inference to show that ¬p\neg p is also false. We can conclude that it was wrong to assume that p is false, so it must be true.
    • Example
      • Let's give a direct proof of the theorem: "There are infinitely many prime numbers"
      • Proof
        • Let's suppose that there are only finitely many prime numbers
        • Let's list them as p1,p2,p3,...,pnp_1, p_2, p_3, ..., p_n where p1=2,p2=3,p3=5p_1 = 2, p_2 = 3, p_3 = 5 and so on.
        • Let's consider the number c = p1 p2 p3...pn+1p_1 \ p_2 \ p_3 ... p_n + 1, the product of all prime numbers, plus 1.
        • Then, as c is a natural number, it has at least one prime divisor.
        • Then k{1...n}\exists k \in \{1 ... n\} where pkcp_k | c (where pkp_k divides cc).
        • Then k{1...n},dN\exists k \in \{1 ... n\}, \exists d \in N where dpk=c=p1 p2 p3...pn+1dp_k = c = p_1 \ p_2 \ p_3 ... p_n + 1
        • Then k{1...n}\exists k \in \{1 ... n\}, dN\exists d \in N where d=p1p2...pk+1...pn+1pkd = p_1 p_2 ... p_{k + 1} ... p_n + \frac{1}{p_k}
        • Then, 1pk\frac{1}{p_k}, in the expression above, is an integer, which is a contradiction.

6.103 The principle of mathematical induction

  • Mathematical Induction
    • Can be used to prove that a propositional function P(n)P(n) is true for all positive integers.
    • The rule of inference: P(1) is trueP(1) \text{ is true} k(P(k)P(k+1))\forall k (P(k) \rightarrow P(k + 1)) nP(n)\therefore \forall n P(n)
      • Intuition:
        • P is true for 1.
        • Since P is true for 1, it's true for 2.
        • Since P is true for 2, it's true for 3.
        • And so on...
        • Since P is true for n-1, it's true for n ...
      • In other words:
        • The base case shows that the property initially holds true.
        • The induction step shows how each iteration influences the next one.
    • Structure of induction
      • In order to prove that a propositional function P(n) is true for all, we need to verify 2 steps:
          1. BASIS STEP: where we show that P(1) is true.
          1. INDUCTION STEP: where we show that for kN\forall k \in \mathbb{N}: if P(k)P(k) is true called inductive hypothesis, then P(k+1)P(k + 1) is true.
    • Some uses of induction
      • Mathematical induction can be used to prove P(n) is true for all integers greater than a particular integer, where P(n) is a propositional function. Might cover multiple cases like:
        • Proving formulas
        • Proving inequalities
        • Proving divisibility
        • Providing properties of subsets and their cardinality.

6.106 Proof by induction

  • Proof By Induction
    • Proving formulas
      • Proving a simple formula formalised as the propositional function, P(n):1+2+3+...+n=n(n+1)/2P(n): 1 + 2 + 3 + ... + n = n (n + 1) / 2
      • The sum of 1 to n is equal to n multipled by n plus 1 divided by 2, for all n in N.
    • In order to prove that a propositional function P(n)P(n) is true for all N, we need to verify 2 step:
        1. BASIS STEP: where we show that P(1) is true.
        1. INDUCTIVE STEP: where we show that for kN\forall k \in \mathbb{N}: if P(k)P(k) is true, called inductive hypothesis, then P(k + 1) is true.
      1. BASIS STEP: The basis step, P(1) reduces to 1 = P(1)=1(2)/2=1P(1) = 1 (2) / 2 = 1
      1. INDUCTIVE STEP:
      2. Let kN\forall k \in \mathbb{N}
      3. If the inductive hypothesis P(k) is true:
        • we have 1+2+3+...+k=k(k+1)/21+2+3+...+k = k(k+1)/2
        • then, 1+2+3+...+k+(k+1)1+2+3+...+k+(k+1)
        • =k(k+1)/2+(k+1)= k(k+1) / 2+(k+1)
        • =(k(k+1)+2(k+1))/2= (k (k + 1) + 2(k + 1)) / 2
        • =(k+1)((k+1)+1)/2= (k + 1) ((k + 1) + 1) / 2
        • which verifies, P(k+1)P(k+1)
  • Proving inequalities
    • We may also use math induction to prove an inequality that holds for all positive integers greater than a particular positive integer.
    • Consider proving the propositional function P(n):3n<n!P(n): 3^n < n! if n is an integer greater than or equal to 7.
        1. BASIS STEP: The basis step, P(7)P(7) reduces to 37<7!3^7 < 7! because 2187 < 5040.
        1. INDUCTIVE STEP:
        2. Let kNk \in \mathbb{N} and k7k \ge 7
        3. If inductive hypothesis P(k)P(k) is true:
          • then 3k+1=33k<(k+1)k!=(k+1)!3^{k+1} = 3 * 3^{k} < (k + 1) * k! = (k + 1)! which verifies P(k+1)P(k + 1) is true.
  • Proving divisibility
    • We can use math induction to prove divisibility that holds for all positive integers greater than positive integer.
    • Consider proving the prop function P(n):nN 6n+4P(n): \forall n \in \mathbb{N} \ 6^{n} + 4 is divisible by 5.
    • Example
        1. BASIS STEP: The basis step, P(0), reduces to 60+46^{0} + 4 is divisible by 5, because 60+4=56^{0} + 4 = 5
        1. INDUCTION STEP:
        2. Let kNk \in \mathbb{N}
        3. If inductive hypothesis P(k) is true:
          • then, 6k+4=5p6^{k} + 4 = 5p where pNp \in \mathbb{N}
          • then, 6k+1+4=6(5p4)+4=30p206{k+1} + 4 = 6 * (5p - 4) + 4 = 30p - 20
          • =5(6p4)= 5(6p - 4) which is divisible by 5 and verifies P(k+1)P(k + 1) is true.
  • Incorrect Induction
    • Consider the statement of the following incorrect induction: P(n):nNi=0n12i=2nP(n): \forall n \in \mathbb{N} \sum^{n-1}_{i=0} 2^{i} = 2^{n}
    • Proof:
      • Let kNk \in \mathbb{N}. Let's suppose the inductive hypothesis P(k)P(k) is true, which means: i=0k12i=2k\sum^{k-1}_{i=0} 2^{i} = 2^{k}
      • Let's examine P(k+1)P(k + 1)
      • i=0k2i=i=0k12i+2k=2k+2k=2k+1\sum^{k}_{i=0} 2^{i} = \sum^{k-1}_{i=0} 2^{i} + 2^{k} = 2^{k} + 2^{k} = 2^{k + 1}
      • This means that P(k+1)P(k + 1) is also true and verifies the induction step.
    • Even though we have been able to prove induction step, let's prove statement: nN\forall n \in \mathbb{N} i=0n12i=2n\sum^{n-1}_{i=0} 2^{i} = 2^{n} is FALSE.
      • For example, 20+21=32^{0} + 2^{1} = 3 which is different from 222^2
    • Our reasoning seemed correct but we didn't verify the base case and have made false assumptions.
    • In other words, as we saw in propositional logic, false assumptions imply false conclusions.
    • To avoid this situation, we need to make sure both the base case and induction step are verified.

6.108 Strong induction

  • Strong Induction
    • Can be formalised with rule of inference:
      • P(1)P(1) is true.
      • kN P(1),P(2)...P(k)P(k+1)\forall k \in \mathbb{N} \ P(1), P(2) ... P(k) \rightarrow P(k+1)
      • nN,P(n)\therefore \forall n \in \mathbb{N}, P(n)
    • Sometimes easier to prove statements using strong induction than other methods.
    • Strong induction is sometimes called: "the 2nd principal of mathematical induction" or "complete induction".
    • Example:
      • Prove propositional function, P(n): nN\forall n \in \mathbb{N} and n2,nn \ge 2, n is divisible by prime number.
      • To prove need to verify 2 steps:
          1. BASIS STEP: P(2)P(2) reduces to 2, which is divisible by prime number because 2 is a prime number and divides itself.
          1. INDUCTIVE STEP:
          2. Let kNk \in \mathbb{N} greater than 2.
          3. If inductive hypothesis, P(k)P(k) is true:
            • Let's also assume P(2)...P(k+1)P(2) ... P(k+1) is true. Then, mN\forall m \in \mathbb{N} and 2mk+1:p2 \le m \le k+1 : \exists p
            • Then, mN\forall m \in \mathbb{N} and 2mk+12 \le m \le k + 1: p\exists p is a prime number dividing m.
            • We have two cases:
              • k + 2 is a prime number, in which case it is trivially divisible by itself.
              • k + 2 is not a prime number, in which case m\exists m dividing k+2k + 2
              • as 2mk+12 \le m \le k + 1, p\exists p is a prime number dividing m. p also divides k+2
              • Which verifies P(k + 2) is true and proves the strong induction.
  • Well-Ordering Property
    • The well-ordering property is an axiom about N\mathbb{N} that we assume to be true. The axioms about N\mathbb{N} are the following:
        1. The number 1 is a positive integer.
        1. If nNn \in \mathbb{N} then n+1n + 1, the successor of n, is also a positive integer.
        1. Every positive integer other than 1 is the successor of a positive integer.
        1. The well-ordering property: every nonempty subset of positive integers has at least one element.
    • The well-ordering property can be used as a tool in building proofs.
    • Example
      • Let's reconsider the earlier statement P(n): nN\forall n \in \mathbb{N} and n2n \ge 2, n is divisible by a prime number.
      • Proof
        • Let SS be the set of positive integers greater than 1 with no prime divisor.
        • Suppose SS is nonempty. Let n be its smallest element.
        • n cannot be a prime, since n divides itself and if n were prime, it would be its own prime divisor.
        • So n is composite: it must have a divisor d with 1 < d < n. Then, d must have a prime divisor d with 1 < d < n. Then, d must have a prime divisor (by the minimality of n), let's call it p.
        • Then p / d and d / n, so p/n, which is a contradiction.
        • Therefore S is empty, which verifies P(n).
  • Equivalence of the three concepts
    • We can prove the following statements:
      • mathematical induction -> the well-ordering property.
      • the well-ordering property -> strong induction.
      • strong induction -> mathematical induction.
    • That is, the principles of mathematical induction, strong induction and well-ordering are all equivalent.
      • The validaity of each of the 3 proof techniques implies the validity of the other 2 technicques.