270 likes | 482 Views
Discrete Mathematics CS 2610. August 21, 2008. Agenda . Nested quantifiers Rules of inference Proofs. Review. A predicate P , or propositional function, is a function that maps objects in the universe of discourse to propositions
E N D
Discrete Mathematics CS 2610 August 21, 2008
Agenda • Nested quantifiers • Rules of inference • Proofs
Review A predicate P, or propositional function, is a function that maps objects in the universe of discourse to propositions • Predicates can be quantified using the universal quantifier (“for all”) or the existential quantifier (“there exists”) • Quantified predicates can be negated as follows • x P(x) x P(x) • x P(x) x P(x) • Quantified variables are called “bound” • Variables that are not quantified are called “free”
Predicate Logic and Propositions • An expression with zero free variables is an actual proposition Ex. Q(x) : x > 0, R(y): y < 10 x Q(x) y R(y)
Nested Quantifiers • When dealing with polyadic predicates, each argument may be quantified with its own quantifier. • Each nested quantifier occurs in the scope of another quantifier. Examples: (L=likes, UoD(x)=kids, UoD(y)=cars) • xy L(x,y) reads x(y L(x,y)) • xy L(x,y) reads x(y L(x,y)) • xy L(x,y) reads x(y L(x,y)) • xy L(x,y) reads x(y L(x,y)) Another example • x (P(x) y R(x,y))
Order matters!!! Examples • If L(x,y) means x likes y, how do you read the following quantified predicates? y L(Alice,y) yx L(x,y) xy L(x,y) x L(x, Prius) Alice likes some car There is a car that is liked by everyone Everyone likes some car Everyone likes the Prius
Negation of Nested Quantifiers • To negate a quantifier, move negation to the right, changing quantifiers as you go. Example: xyz P(x,y,z) x y z P(x,y,z).
Proofs and inference Assume that the following statements are true: I have a total score over 96. If I have a total score over 96, then I get an A in the class. What can we claim? I get an A in the class. How do we know the claim is true? Logical Deduction.
Proofs • A theorem is a statement that can be proved to be true. • A proof is a sequence of statements that form an argument.
Proofs: Inference Rules • An Inference Rule: “” means “therefore” premise 1 premise 2 … conclusion
p p q q Proofs: Modus Ponens I have a total score over 96. If I have a total score over 96, then I get an A for the class. I get an A for this class Tautology: (p (p q)) q
q p q p Proofs: Modus Tollens • If the power supply fails then the lights go out. • The lights are on. The power supply has not failed. Tautology: (q (p q)) p
p p q Proofs: Addition • I am a student. I am a student or I am a visitor. Tautology: p (p q)
p q p Proofs: Simplification • I am a student and I am a soccer player. I am a student. Tautology: (p q) p
p q p q Proofs: Conjunction • I am a student. • I am a soccer player. I am a student and I am a soccer player. Tautology: ((p) (q)) p q
p q q p Proofs: Disjunctive Syllogism I am a student or I am a soccer player. I am a not soccer player. I am a student. Tautology: ((p q) q) p
p q q r p r Proofs: Hypothetical Syllogism If I get a total score over 96, I will get an A in the course. If I get an A in the course, I will have a 4.0 semester average. If I get a total score over 96 then • I will have a 4.0 semester average. Tautology: ((p q) (q r)) (p r)
p q p r q r Proofs: Resolution I am taking CS1301 or I am taking CS2610. I am not taking CS1301 or I am taking CS 1302. I am taking CS2610 or I am taking CS 1302. Tautology: ((p q ) ( p r)) (q r)
p q p r q r r Proofs: Proof by Cases I have taken CS2610 or I have taken CS1301. If I have taken CS2610 then I can register for CS2720 If I have taken CS1301 then I can register for CS2720 I can register for CS2720 Tautology: ((p q ) (p r) (q r)) r
q p q p Fallacy of Affirming the Conclusion • If you have the flu then you’ll have a sore throat. You have a sore throat. You must have the flu. Fallacy: (q (p q)) p Abductive, rather than deductive reasoning!
p p q q Fallacy of Denying the Hypothesis • If you have the flu then you’ll have a sore throat. • You do not have the flu. You do not have a sore throat. Fallacy: (p (p q)) q
Inference Rules for Quantified Statements Universal Instantiation (for an arbitrary object c from UoD) xP(x)P(c) Universal Generalization (for any arbitrary element c from UoD) P(c)___ xP(x) xP(x)P(c) Existential Instantiation (for some specific object c from UoD, that has not yet occurred!) Existential Generalization (for some specific object c from UoD) P(c)__ xP(x)
Vacuous & Trivial Proofs p q is vacuously true if p is false In this case, p q is a vacuous proof p q is trivially true if q is true In this case, we have a trivial proof
Proofs • Direct Proof: To prove p q, we assume p and show/derive q • Indirect Proof by Contraposition: To prove p q, we prove its contrapositive, q p • So assume q and show/derive p
Proofs Indirect Proof by Contradiction: To prove p, we assume p and derive a contradiction. Based on the tautology ( p F ) p “if the negation of p implies a contradiction then p must be true” (aka: Reductio ad Absurdum)
More Proofs • Equivalence: To prove p q, we prove p q and q p • Cases: To prove (p1 v p2 v … v pn) q, we prove (p1 q) (p2 q) … (pn q)
More Proofs Quantifiers: x P(x) : provide a proof, or counterexample. x P(x): Existence Constructive Proof: Find an a in the UoD such that P(a) holds. Existence Non-Constructive Proof: Prove that x P(x) is true without finding an a in the UoD such that P(a) holds