Mathematical Logic Part Three

Transcription

1 Mathematical Logic Part Three

2 Recap from Last Time

3 What is First-Order Logic? First-order logic is a logical system for reasoning about properties of objects. Augments the logical connectives from propositional logic with predicates that describe properties of objects, and functions that map objects to one another, quantifiers that allow us to reason about multiple objects simultaneously.

4 For any natural number n, n is even iff n2 is even n. (n ℕ (Even(n) Even(n2))) isis the the universal universalquantifier quantifier and and says says for for any any choice choice of of n, n, the the following following isis true. true.

5 Some muggles are intelligent. m. (Muggle(m) Intelligent(m)) isis the the existential existentialquantifier quantifier and and says says for for some some choice choice of of m, m, the the following following isis true. true.

6 All P's are Q's translates as x. (P(x) Q(x))

7 Useful Intuition: Universally-quantified statements are true unless there's a counterexample. x. (P(x) Q(x)) If If xx isis aa counterexample, counterexample, itit must must have have property property PP but but not not have have property property Q. Q.

8 Some P is a Q translates as x. (P(x) Q(x))

9 Useful Intuition: Existentially-quantified statements are false unless there's a positive example. x. (P(x) Q(x)) If If xx isis an an example, example, itit must must have have property property PP on on top top of of property property Q. Q.

10 Good Pairings The quantifier usually is paired with. The quantifier usually is paired with. In the case of, the connective prevents the statement from being false when speaking about some object you don't care about. In the case of, the connective prevents the statement from being true when speaking about some object you don't care about.

11 The Art of Translation

12 Using the predicates - Person(p), which states that p is a person, and - Loves(x, y), which states that x loves y, write a sentence in first-order logic that means everybody loves someone else.

13 Everybody loves someone else

14 Every person loves some other person

15 Every person p loves some other person

16 p. (Person(p) p loves some other person )

17 p. (Person(p) there is some other person that p loves )

18 p. (Person(p) there is a person other than p that p loves )

19 p. (Person(p) there is a person q, other than p, where p loves q )

20 p. (Person(p) there is a person q, other than p, where p loves q )

21 p. (Person(p) q. (Person(q), other than p, where p loves q ) )

22 p. (Person(p) q. (Person(q) p q p loves q ) )

23 p. (Person(p) q. (Person(q) p q Loves(p, q) ) )

24 Using the predicates - Person(p), which states that p is a person, and - Loves(x, y), which states that x loves y, write a sentence in first-order logic that means there is a person that everyone else loves.

25 There is a person that everyone else loves

26 There is a person p where everyone else loves p

27 p. (Person(p) everyone else loves p )

28 p. (Person(p) every other person q loves p )

29 p. (Person(p) every person q, other than p, loves p )

30 p. (Person(p) q. (Person(q) p q q loves p )

31 p. (Person(p) q. (Person(q) p q Loves(q, p) ) )

32 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else.

33 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else. p. (Person(p) q. (Person(q) p q Loves(p, q)))

34 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else. p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person,

35 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else. p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person, there is some person

36 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else. p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person, there is some person who isn't them

37 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: Everyone loves someone else. p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person, there is some person who isn't them that they love.

38 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves.

39 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves. p. (Person(p) q. (Person(q) p q Loves(q, p)))

40 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person

41 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person who everyone

42 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person who everyone who isn't them

43 Combining Quantifiers Most interesting statements in first-order logic require a combination of quantifiers. Example: There is someone everyone else loves. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person who everyone who isn't them loves.

44 For Comparison p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person, there is some person who isn't them that they love. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person who everyone who isn't them loves.

45 Everyone Loves Someone Else

46 Everyone Loves Someone Else No No one one here here isis universally universally loved. loved.

47 There is Someone Everyone Else Loves

48 There is Someone Everyone Else Loves This This person person does does not not love love anyone anyone else. else.

49 Everyone Loves Someone Else and There is Someone Everyone Else Loves

50 p. (Person(p) q. (Person(q) p q Loves(p, q))) For every person, there is some person who isn't them that they love. p. (Person(p) q. (Person(q) p q Loves(q, p))) There is some person who everyone who isn't them loves.

51 Quantifier Ordering The statement x. y. P(x, y) means for any choice x, there's some y where P(x, y) is true. The choice of y can be different every time and can depend on x.

52 Quantifier Ordering The statement x. y. P(x, y) means there is some x where for any choice of y, we get that P(x, y) is true. Since the inner part has to work for any choice of y, this places a lot of constraints on what x can be.

53 Order matters when mixing existential and universal quantifiers!

54 Time-Out for Announcements!

55 Problem Set Two Problem Set One was due at 3:00PM today. You can submit late up until Monday at 3:00PM. Problem Set Two goes out now. Checkpoint due Monday. Remaining problems due next Friday. Play around with propositional logic, firstorder logic, and their applications! As always, feel free to ask us questions!

56 Global Engineering Programs Fair Stanford's Global Engineer Programs office (GEP) is holding a Global Engineering Programs Fair. Representatives will be there from Bing Overseas Studies, the Global Studies Department, and the Overseas Resource Center. October 13 from 4PM 6PM in Huang Amphitheater. Questions? ming@stanford.edu.

57 WiCS Industry Mentorship Stanford WiCS (Women in Computer Science) runs an industry mentorship program pairing Stanford students with industry professionals. Interested? Apply online by October 15:

58 RSVP Here

59 Picking Up Solution Sets

60

61

62 This is the CS103 drawer This is the CS103A drawer

63 Your Questions

64 What was your first career fair and internship like? Oh man, I made myself look like an idiot on many Oh man, I made myself look like an idiot on many occasions at that career fair. No one told me how to occasions at that career fair. No one told me how to prepare or how to write a good resume, which is one of prepare or how to write a good resume, which is one of the reasons I'm happy we offer CS9. the reasons I'm happy we offer CS9. My first internship was interesting. I was working at My first internship was interesting. I was working at Mozilla, so I could show everyone my code. I don't think Mozilla, so I could show everyone my code. I don't think it was the best fit for me, but I only figured that out it was the best fit for me, but I only figured that out afterwards. Still, it was a fun experience! afterwards. Still, it was a fun experience!

65 Why do some PhDs do research in industry over academia? Replace some with most. Replace some with most. People tend to go where the interesting problems are. Many People tend to go where the interesting problems are. Many companies work on problems that are at a large scale or that have a companies work on problems that are at a large scale or that have a huge impact. Companies also often pay much better and don't have huge impact. Companies also often pay much better and don't have a demanding tenure process. a demanding tenure process. That said, academia has its advantages. You get to mentor new That said, academia has its advantages. You get to mentor new students, work on fundamental research or highly theoretical topics, students, work on fundamental research or highly theoretical topics, and (post-tenure) have incredible job security. and (post-tenure) have incredible job security.

66 How did you and your girlfriend meet? Thanks for asking, but this Thanks for asking, but this one's a bit personal and one's a bit personal and I'd prefer not to share. I'd prefer not to share.

67 What private Silicon Valley tech companies do you find most promising or fascinating? Tesla and SpaceX are doing some amazing work, though I've Tesla and SpaceX are doing some amazing work, though I've heard that the work is super demanding. Some of the ML heard that the work is super demanding. Some of the ML work that Google is doing is truly amazing. Khan Academy has work that Google is doing is truly amazing. Khan Academy has huge impact and is a great place to work (or intern!). huge impact and is a great place to work (or intern!). Palantir makes some really neat technology as well, though Palantir makes some really neat technology as well, though make sure you're okay with what they do with it. make sure you're okay with what they do with it.

68 Back to CS103!

69 Set Translations

70 Using the predicates - Set(S), which states that S is a set, and - x y, which states that x is an element of y, write a sentence in first-order logic that means the empty set exists.

71 Using the predicates - Set(S), which states that S is a set, and - x y, which states that x is an element of y, write a sentence in first-order logic that means the empty set exists. First-order First-order logic logic doesn't doesn't have have set set operators operators or or symbols symbols built built in. in. If If we we only only have have the the predicates predicates given given above, above, how how might might we we describe describe this? this?

72 The empty set exists.( ( (

73 There is some set S that is empty.( ( (

74 S. (Set(S) S is empty. )

75 S. (Set(S) there are no elements in S )

76 S. (Set(S) there is an element in S )

77 S. (Set(S) there is an element x in S )

78 S. (Set(S) x. x S )

79 S. (Set(S) x. x S)

80 S. (Set(S) x. x S) S. (Set(S) there are no elements in S( )

81 S. (Set(S) x. x S) S. (Set(S) every object does not belong to S( )

82 S. (Set(S) x. x S) S. (Set(S) every object x does not belong to S( )

83 S. (Set(S) x. x S) S. (Set(S) x. (x S) )

84 S. (Set(S) x. x S) S. (Set(S) x. (x S))

85 S. (Set(S) x. x S) S. (Set(S) x. (x S)) Both Both of of these these translations translations are are correct. correct. Just Just like like inin propositional propositional logic, logic, there there are are many many different different equivalent equivalent ways ways of of expressing expressing the the same same statement statement inin firstfirstorder order logic. logic.

86 Using the predicates - Set(S), which states that S is a set, and - x y, which states that x is an element of y, write a sentence in first-order logic that means two sets are equal if and only if they contain the same elements.

87 Using the predicates - Set(S), which states that S is a set, and - x y, which states that x is an element of y, write a sentence in first-order logic that means two sets are equal if and only if they contain the same elements. This This isis called called the the axiom axiom of of extensionality, extensionality, by by the the way. way.

88 Two sets are equal if and only if they have the same elements. ) ) ) ) )

89 Any two sets are equal if and only if they have the same elements.) ) ) ) )

90 Any two sets S and T are equal if and only if they have the same elements.) ) ) ) )

91 S. (Set(S) T. (Set(T) ) S and T are equal if and only if they have the same ) elements.) ) )

92 S. (Set(S) T. (Set(T) (S = T if and only if they have the same elements.)) ) )

93 S. (Set(S) T. (Set(T) (S = T they have the same elements.)) ) )

94 S. (Set(S) T. (Set(T) (S = T S and T have the same elements.)) ) )

95 S. (Set(S) T. (Set(T) (S = T every element of S is an element of T and vice-versa) ) )

96 S. (Set(S) T. (Set(T) (S = T x is an element of T if and only if x is an element of S) ) )

97 S. (Set(S) T. (Set(T) (S = T x. (x S x T)) ) )

98 S. (Set(S) T. (Set(T) (S = T x. (x S x T)) ) )

99 S. (Set(S) T. (Set(T) (S = T x. (x S x T)) ) ) You You sometimes sometimes see see the the universal universal quantifier quantifier pair pair with with the the connective. connective. This This isis especially especially common common when when talking talking about about sets sets because because two two sets sets are are equal equal when when they they have have precisely precisely the the same same elements. elements.

100 Mechanics: Negating Statements

101 Negating Quantifiers We spent much of Monday's lecture discussing how to negate propositional constructs. How do we negate statements with quantifiers in them?

102 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) For some choice of x, P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

103 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) For some choice of x, P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

104 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

105 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

106 An Extremely Important Table x. P(x) When is this true? When is this false? For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, For any choice of x, P(x) P(x) x. P(x) For any choice of x, For some choice of x, P(x) P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x)

107 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) x. P(x) x. P(x) For any choice of x, For some choice of x, P(x) P(x) For some choice of x, P(x) For any choice of x, P(x)

108 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

109 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) For some choice of x, P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

110 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

111 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) x. P(x) For some choice of x, P(x) For any choice of x, P(x) x. P(x) x. P(x)

112 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) x. P(x) x. P(x) x. P(x) For some choice of x, For any choice of x, P(x) P(x)

113 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) x. P(x) For some choice of x, P(x) x. P(x) x. P(x) x. P(x)

114 An Extremely Important Table When is this true? When is this false? x. P(x) For any choice of x, P(x) x. P(x) x. P(x) For some choice of x, P(x) x. P(x) For any choice of x, P(x) x. P(x) For some choice of x, P(x) x. P(x) x. P(x) x. P(x)

115 Negating First-Order Statements Use the equivalences x. φ x. φ x. φ x. φ to negate quantifiers. Mechanically: Push the negation across the quantifier. Change the quantifier from to or vice-versa. Use techniques from propositional logic to negate connectives.

116 Taking a Negation x. y. Loves(x, y) ( Everyone loves someone. ) x. y. Loves(x, y) x. y. Loves(x, y) x. y. Loves(x, y) ( There's someone who doesn't love anyone. )

117 Two Useful Equivalences The following equivalences are useful when negating statements in first-order logic: (p q) p q (p q) p q These identities are useful when negating statements involving quantifiers. is used in existentially-quantified statements. is used in universally-quantified statements. When pushing negations across quantifiers, we strongly recommend using the above equivalences to keep with and with.

118 Negating Quantifiers What is the negation of the following statement, which says there is a cute puppy? x. (Puppy(x) Cute(x)) We can obtain it as follows: x. (Puppy(x) Cute(x)) x. (Puppy(x) Cute(x)) x. (Puppy(x) Cute(x)) This says every puppy is not cute. Do you see why this is the negation of the original statement from both an intuitive and formal perspective?

119 S. (Set(S) x. (x S)) ( There is a set that doesn't contain anything ) S. (Set(S) x. (x S)) S. (Set(S) x. (x S)) S. (Set(S) x. (x S)) S. (Set(S) x. (x S)) S. (Set(S) x. x S) ( Every set contains at least one element )

120 These two statements are not negations of one another. Can you explain why? S. (Set(S) x. (x S)) ( There is a set that doesn't contain anything ) S. (Set(S) x. (x S)) ( Everything is a set that contains something ) Remember: Remember: usually usually goes goes with with,, not not

121 Restricted Quantifiers

122 Quantifying Over Sets The notation x S. P(x) means for any element x of set S, P(x) holds. This is not technically a part of first-order logic; it is a shorthand for x. (x S P(x)) How might we encode this concept? x S. P(x) Answer: x. (x S P(x)). Note Note the the use use of of instead instead of of here. here.

123 Quantifying Over Sets The syntax x S. φ x S. φ is allowed for quantifying over sets. In CS103, feel free to use these restricted quantifiers, but please do not use variants of this syntax. For example, don't do things like this: x with P(x). Q(x) y such that P(y) Q(y). R(y).

124 Expressing Uniqueness

125 Using the predicate - Level(l), which states that l is a level, write a sentence in first-order logic that means there is only one level. AAfun fundiversion: diversion:

126 There is only one level.

127 Something is a level, and nothing else is.

128 Some thing l is a level, and nothing else is.

129 Some thing l is a level, and nothing besides l is a level

130 l. (Level(l) nothing besides l is a level. )

131 l. (Level(l) anything that isn't l isn't a level )

132 l. (Level(l) any thing x that isn't l isn't a level )

133 l. (Level(l) x. (x l x isn't a level) )

134 l. (Level(l) x. (x l Level(x)) )

135 Expressing Uniqueness To express the idea that there is exactly one object with some property, we write that there exists at least one object with that property, and that there are no other objects with that property. You sometimes see a special uniqueness quantifier used to express this:!x. φ For the purposes of CS103, please do not use this quantifier. We want to give you more practice using the regular and quantifiers.

136 Next Time Binary Relations First-Order Definitions Equivalence Relations Strict Orders