0% found this document useful (0 votes)
114 views

Logic Set

This document provides revision notes and problems for logic and set theory. It covers topics like propositions, logic operators, truth tables, equivalence, contrapositive, predicates, quantifiers, proofs, sets, relations, functions, and cardinality. The notes are intended as a study guide for students and do not replace in-class lectures. It contains 10 chapters with examples and exercises to help students review key concepts.

Uploaded by

Lucian Guzgan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
114 views

Logic Set

This document provides revision notes and problems for logic and set theory. It covers topics like propositions, logic operators, truth tables, equivalence, contrapositive, predicates, quantifiers, proofs, sets, relations, functions, and cardinality. The notes are intended as a study guide for students and do not replace in-class lectures. It contains 10 chapters with examples and exercises to help students review key concepts.

Uploaded by

Lucian Guzgan
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 24

LOGIC

&
SET THEORY
AMIN WITNO
..
www. wi t n o . c o m
Logic & Set Theory
Revision Notes and Problems
Amin Witno <www.witno.com>
Pre!ce
These notes are for students of Math 25 as a revision wor!boo! and are not meant to
substitute the in"class notes. No student is e#$ected to reall% benefit from these notes
unless the% have re&ularl% attended the lectures.
'ha$ter ( Pre"imin!rie#
The Real Numbers and )ts *ubsets+ )nterval Notations+ Absolute ,alues+ Modulo
-$erations+ *e.uences+ *i&ma Notations
'ha$ter Logic
Pro$ositions+ /o&ic -$erators+ Truth Tables+ 0.uivalence+ 'ontra$ositive+ Predicates
and 1uantifiers
'ha$ter 2 Proo#
Provin& 'onditional *tatements+ Proof b% 'ontra$ositive+ Provin& 0.uivalence
*tatements+ Proof b% 'ases+ Proof b% 'ontradiction+ Provin& 0#istence+ Provin&
2ni.ueness+ Provin& Not"All *tatements+ The Princi$les of Mathematical )nduction
'ha$ter 3 Set#
*et -$erations+ ,enn 4ia&rams+ *et )dentities+ *ubsets+ Power *et+ 'ardinalit%+ 'ross
Product+ 5enerali6ed 2nions and )ntersections
'ha$ter 7 Re"!tion#
Relations on a *et+ )nverse and 'om$ositions+ 4i&ra$hs+ 0.uivalence Relations and
0.uivalence 'lasses+ Partial -rder Relations+ 8asse 4ia&rams+ Total -rderin&+ Well
-rderin&+ The Well -rderin& Princi$le+ 9ero"-ne Matrices+ Transitive 'losures
'ha$ter 5 $%nction#
-ne"to"one :unctions+ -nto :unctions+ )nverse and 'om$ositions+ ;i<ections
'ha$ter = C!r&in!"ity
'ountable *ets+ 'antor"*chroeder";ernstein Theorem+ 2ncountable *ets
Reerence#
. *mith+ 0&&en+ and *t. Andre+ A Transition to Advanced Mathematics+ >
th
edition
2(+ ;roo!s 'ole.
2. 4. *olow+ 8ow to Read and 4o Proofs? An )ntroduction to Mathematical Thou&ht
Processes+ 5
th
edition 2((@+ Wile%.
3. Michael /. -A/ear%+ The *tructure of Proof with /o&ic and *et Theor%+ 2((2+
Prentice 8all.
7. A. Witno+ 4iscrete *tructures in :ive 'ha$ters+ 2((+ 'reate*$ace.
Co'yright#
'o$%ri&hted under a 'reative 'ommons /icense.
B 2((=C2(( Amin Witno /ast 0dited? @C2C2((

'ha$ter (
Pre"imin!rie#
)n mathematics ver% often we stud% sets whose elements are the real numbers. *ome
s$ecial number sets which are fre.uentl% encountered are defined as follow.
The set of natural numbers contains the elements + 2+ 3+ ...
The set of inte&ers contains all the natural numbers to&ether with their
ne&atives and 6ero? ... + C3+ C2+ C+ (+ + 2+ 3+ ...
The set of rational numbers consists of numbers of the form a/b where a and
b are inte&ers with b D (+ for e#am$les E2+ 5 F 5E+ C22E>+ (E@+ etc. 8ence all
inte&ers are rational numbers+ but some rational numbers are not inte&ers.
The set of all real numbers is denoted b% .
The set of irrational numbers consists of all real numbers which are not
rational+ such as G2+ H+ etc.
The set of even numbers contains the elements (+ I2+ I7+ I=+ ... which are
those of the form 2n for some inte&er n.
The set of odd numbers is the set of inte&ers which are not even. 8ence odd
numbers are I+ I3+ I5+ ... which can be written as 2n J for some inte&er n.
(. Prove that the number G2 is not rational.
A set of real numbers x in the ran&e a < x < b can also be written usin& the inter(!"
not!tion Ka, bL. The round brac!et at either end can be re$laced b% a s.uare brac!et to
indicate inclusion. :or e#am$le Ka, bM means the set a < x b. Moreover we use the
infinit% s%mbol to indicate unboundedness+ such as Na+ OL for the set x a.
(.2 Write the interval notation for each set.
aL a x < b
bL a x b
cL x < b
dL x b
:or real numbers x we define the !)#o"%te (!"%e of x to be |x| = x if x ( and |x| = x
if x < (. :or e#am$le PC2P F 2+ PG2P F G2+ and P(P F (. A useful fact is that GKx
2
L F |x|.
(.3 :ind all real number solutions of these e.uations.
aL PxP F 3
bL Px J P F 3
cL Px J P > 3
dL P2x J P > 3
:or real numbers x, the gre!te#t integer %nction NxM &ives the &reatest inte&er not
&reater than x. :or e#am$le N3.7M F 3.
(.7 0valuate NxM for these values of x.
aL 5
bL .@@@
2
cL 237E5
dL C2.3
eL G(
:or two inte&ers m and n > ( define the mo&%"o o'er!tion m mod n = m C Nm/nM n.
:or e#am$le 2>E5 F 73.7 hence 2> mod 5 F 2> C K73 Q 5L F 2. 0.uivalentl% 2> F
K73L Q 5 J K2L hence 2> mod 5 F 2+ which is the remainder when 2> is divided b% 5.
(.5 0valuate the followin&.
aL 23 mod 3
bL 2((( mod >
cL 25 mod 5
dL 25 mod
eL mod 25
Note that m mod n is the remainder when m is divided b% n. )n $articular m mod n F (
when m is a m%"ti'"e of n, or we sa% that n &i(i&e# m. :or e#am$le 2 mod 3 F (
because 2 F 3 Q 7+ so we sa% 3 divides 2. Also m mod 2 F ( whenever m is an even
number+ so all even numbers are multi$les of 2.
A #e*%ence is a function f KnL defined over the natural numbers+ hence it can be
ordered as f KL+ f K2L+ f K3L+ ...
0#am$les? L f KnL F n
2
is the se.uence + 7+ @+ =+ 25+ 3=+ 7@+ ...
2L f KnL F 2n C is the se.uence + 3+ 5+ >+ @+ + 3+ ...
(.= Write out the followin& se.uences.
aL f KnL F 2n J
bL nKn J L
cL n mod 5
dL NnE2M
(.> :ind a formula f KnL for each se.uence.
aL + 2+ 7+ R+ =+ 32+ =7+ ...
bL 3+ =+ @+ 2+ 5+ R+ 2+ ...
cL >+ + 5+ @+ 23+ 2>+ 3+ ...
dL + 2+ 3+ + 2+ 3+ + 2+ 3+ ...
*ummations over some or all terms in a se.uence can be re$resented usin& #igm!
not!tion. :or e#am$le

n=
5
n
2
=7@=25 .
(.R Write the followin& summations usin& si&ma notations.
aL = J 32 J =7 J 2R
bL 2 J 7 J = J R J ( J ...
cL 3 J = J @ J 2 J ... J 3((
dL J 3 J 5 J > J @ J ...
3
'ha$ter
Logic
A 'ro'o#ition is a statement which has a truth value either true or false. :or
e#am$les+ S2 is evenT+ S2 J 2 F 7T+ S2 J 2 F 5T.
The neg!tion of a $ro$osition p is also called not p, and is denoted b% Up.
0#am$le? L )f p? S2 is evenT then Up? S2 is not evenT.
2L )f p? S2 J 2 F 5T then Up? S2 J 2 D 5T.
)f p and q are two $ro$ositions then their con+%nction is the $ro$osition whose value
is true onl% when both are true. A con<unction can also be written p q which is read
p !n& q.
. /et p? S2 is evenT and q? S2 J 2 F 5T. *tate these $ro$ositions and find their value.
aL p q
bL p Uq
cL Up q
dL Up Uq
*imilarl% the &i#+%nction of p and q has value false onl% when both are false. )t is
denoted b% p q and read p or q.
.2 Re$eat Problem . with re$laced b% .
The im'"ic!tion of p and q has value false onl% when p is true and q is false. )t is
denoted b% p V q and read i p then q. A statement in the form p V q is also called a
con&ition!" #t!tement, in which p is a #%icient condition for q and q is a nece##!ry
condition for p.
.3 Re$eat Problem . with re$laced b% V.
The e*%i(!"ence #t!tement p W q is true onl% when p and q have the same value. )t
is read p i !n& on"y i q and is also called a )icon&ition!" #t!tement+ in which p is a
necessar% and sufficient condition for q, and vice versa.
.7 Re$eat Problem . with re$laced b% W.
.5 /et p? SToda% is coldT+ q? SToda% is hotT+ and r? SToda% is wind%T. Write the
followin& $ro$ositions usin& p, q, r.
aL Toda% is hot if and onl% if not wind%.
bL 0ither toda% is cold or not cold.
cL )f toda% is not wind% then it is not hot.
dL Toda% is neither cold nor wind%.
eL )f toda% is wind% then either it is hot or cold.
7
/o&ic o$erators can be $resented in their tr%th t!)"e#?
p q p q p q p - q p . q
T T T T T T
T $ $ T $ $
$ T $ T T $
$ $ $ $ T T
.= 4raw the truth table for each of the followin& $ro$ositions.
aL Up Uq
bL UKp qL V p
cL Kp UqL W KUp qL
dL Kp V qL V r
eL NKp qL V rM W NUp Kq W UrLM
Two $ro$ositions are e*%i(!"ent if their truth tables are identical. We write p q
when the two are e.uivalent. :or e#am$le we can show that Up Uq UKp qL.
.> Prove the followin& e.uivalences b% drawin& the truth tables.
aL Up Uq UKp qL
bL p V q p q
cL p W q Kp V qL Kq V pL
dL p V Kq V rL q V Kp V rL
The contr!'o#iti(e of p V q is the $ro$osition Uq V Up. )t can be shown that these
two are e.uivalent? p V q Uq V Up.
.R Write an e.uivalent statement usin& contra$ositive.
aL )f ) stud% hard then ) &et &ood mar!.
bL )f it rains then it is not hot.
cL )f toda% is not *unda% then tomorrow is not Monda%.
dL )f ) am not la6% then ) come to the lecture.
The con(er#e of p V q is the $ro$osition q V p.
.@ Write the converse of the $ro$ositions in Problem .R. )s p V q q V pX
Theorem? The followin& is a list of some common lo&ical e.uivalence rules?
L p q q p
p q q p
2L p Kq rL Kp qL r
p Kq rL Kp qL r
3L p Kq rL Kp qL Kp rL
p Kq rL Kp qL Kp rL
7L UKUpL p
(p q) p q
UKp qL Up Uq
5L p V q p q
p W q Kp V qL Kq V pL
5
.( Prove b% a$$l%in& the above rules.
aL Kp V qL p q
bL p V q q V p
cL p V Kq V rL q V Kp V rL
dL p V Kq rL Kp V qL Kp V rL
eL Kp

qL V r Kp V rL Kq V rL
. True or :alse. Prove b% an% method %ou li!e.
aL p V Kq V rL Kp V qL V r
bL p V Kq rL Kp V qL Kp V rL
cL p Kq V rL Kp qL V Kp rL
dL p Kq V rL Kp qL V Kp rL
A 're&ic!te is a $ro$ositional function such as PKxL? x J 2 F 5. The statement Sx J 2 F
5T b% itself is not a $ro$osition because it does not have a truth value. ;ut for each
value of x, PKxL becomes a $ro$osition+ for instance+ PK3L? 3 J 2 F 5 is true and PK2L?
2 J 2 F 5 is false.
.2 /et PKxL? x
2
< x.
aL What is the value of PKLX
bL What is the value of PK2LX
cL :or which x is the value of PKxL trueX
.3 /et PKx,yL? x
2
J y
2
F Kx J yL
2
. :ind the values of the followin& $ro$ositions.
aL PK(+L
bL PK(+(L
cL PK+L
dL :or which Kx,yL is the value of PKx,yL trueX
A $redicate can also be made a $ro$osition b% addin& a *%!ntiier. There are three
.uantifiers?
L ? for all E for an% E for each E for ever%
2L ? for some E there is E there e#ists E there is at least one
3L Y ? there is a uni.ue E there is e#actl% one E there e#ists onl% one
0#am$le? /et PKxL? x J 2 F 5 .
L x PKxL? Sfor all real numbers x, x J 2 F 5T+ which is false.
2L x PKxL? Sthere is a real number x such that x J 2 F 5T+ which is true.
3L Y x PKxL? Sthere is a uni.ue real number x such that x J 2 F 5T+ which
is true.
.7 /et PKxL? x < 2x.
aL What is the value of x PKxLX
bL What is the value of x PKxLX
cL What is the value of Y x PKxLX
=
.5 /et PKx, yL? x
2
J y
2
F Kx J yL
2
. :ind the values of the followin& $ro$ositions.
aL x y PKx, yL
bL x y PKx, yL
cL x y PKx, yL
dL y x PKx, yL
eL y x PKx, yL
.= Re$eat Problem .5+ em$lo%in& Y instead of .
.> Re$eat Problem .5 usin& the followin& $redicates.
aL PKx, yL? x
2
J y
2
> (
bL PKx, yL? x
2
J y
2
Z
cL PKx, yL? x
2
C y
2
Z (
dL PKx, yL? x
2
C y > (
We observe+ at least intuitivel%+ that the ne&ations of and are correlated in the
followin& manner.
Ux PKxL y PKxL
Ux PKxL y PKxL
0#am$le? /et PKxL? x J 2 F 5 .
L x PKxL? Sthere is a real number x such that x J 2 F 5T.
Ux PKxL? Sthere is no real number x such that x J 2 F 5T which is
e.uivalent to y PKxL? Sfor all real numbers x, x J 2 D 5T.
2L x PKxL? Sfor all real numbers x, x J 2 F 5T.
Ux PKxL? Snot all real numbers x satisfies x J 2 F 5T which is
e.uivalent to y PKxL? Sthere is a real number x such that x J 2 D 5T.
.R Write the ne&ations b% interchan&in& and .
aL There is a real number x such that x
2
< (.
bL 0ver% inte&er is even.
cL All trian&les have an&le sum e.ual R( de&rees.
dL There is an inte&er x such that x
2
J 2x J 3 F (.
.@ What is the ne&ation of Y x PKxLX 2se %our answer to write the ne&ation of the
statement SThere is a uni.ue real number x such that Ax
2
J Bx J C F (T.
>
'ha$ter 2
Proo#
Pro(ing Con&ition!" St!tement#?
To $rove a $ro$osition in the form p V q, we be&in b% assumin& that p is
true and then show that q must be true.
0#am$le? Prove that if x is an odd inte&er then x
2
is also odd.
*olution? /et p? x is odd+ and q? x
2
is odd. We want to $rove p V q.
*tart? p? x is odd
V x F 2n J for some inte&er n
V x
2
F K2n J L
2
V x
2
F 7n
2
J 7n J
V x
2
F 2K2n
2
J 2nL J
V x
2
F 2m J + where m F K2n
2
J 2nL is an inte&er
V x
2
is odd
V q
2. Prove the followin& $ro$ositions.
aL )f x is an even number then x
3
is also even.
bL )f x is odd then x
2
C 3x is even.
cL )f x and y are odd then x J y is even.
dL )f x and y are even then 7 divides xy.
eL )f x is odd then x
2
C is a multi$le of R.
Proo )y Contr!'o#iti(e?
To $rove a $ro$osition in the form p V q we ma% instead $rove its
contra$ositive? q V p. This wor!s because p V q q V p.
0#am$le? Prove that if x
2
is odd then x must be odd.
*olution? /et p? x
2
is odd+ and q? x is odd. We $rove p V q b% $rovin& q V p.
*tart? q? x is even
V x F 2n for some inte&er n
V x
2
F K2nL
2
V x
2
F 7n
2
V x
2
F 2K2n
2
L
V x
2
F 2m, where m F 2n
2
is an inte&er
V x
2
is even
V p
2.2 Prove the followin& $ro$ositions.
aL )f x
2
is even then x must be even.
bL )f x
3
is even then x must be even.
R
cL )f x
2
C 2x is even then x is even.
dL )f x
3
C 7x J 2 is odd then x is odd.
Pro(ing E*%i(!"ent St!tement#?
To $rove a $ro$osition in the form p W q we must $rove both p V q and
its converse q V p. This is so because p W q Kp V qL Kq V pL.
0#am$le? Prove that x is odd if and onl% if x
2
is odd.
*olution? /et p? x is odd+ and q? x
2
is odd. We must $rove p V q as well as q V p.
;oth of these have been shown in the $revious two e#am$les.
2.3 Prove the followin& $ro$ositions.
aL x is even if and onl% if x
2
is even.
bL x
3
is even if and onl% if x

is even.
cL xy is odd if and onl% if both x and y are odd.
dL x
3
J x
2
J x J is even if and onl% if x is odd.
2.7 Prove that a mod n F b mod n if and onl% if n divides Ka bL.
Proo )y C!#e#?
To $rove a $ro$osition in the form p V q where p a

b we ma% instead
$rove both a V q and b V q.
2.5 Prove the e.uivalence Ka

bL V q Ka V qL Kb V qL.
0#am$le? Prove that if x is an inte&er then x
2
J x is even.
*olution? /et p: x is inte&er+ and q? x
2
J x is even. We must $rove p V q.
/et a? x is even+ and b? x is odd. Then p a

b because an% inte&er is
either even or odd. We will now $rove the two cases a V q and b V q...
2.= Prove the followin& $ro$ositions.
aL )f x

is an inte&er then x
2
C x is even.
bL )f x or y is even then xy is even.
cL )f x

is an inte&er then x
2
J 2 is not a multi$le of 7.
dL )f # is a real number then CPxP [ x [ PxP.
Proof b% cases can be &enerali6ed to three Kor moreL ste$s. *u$$ose we want to $rove
p V q where p a

b c. Then we must $rove the three cases a V q and b V q
and c V q.
2.> Prove that if x

is an inte&er then x
3
C x is a multi$le of 3. 2se the fact that ever%
inte&er comes in the form 3n J k+ where k F ( or or 2.
2.R Prove that if x

and y are real numbers then |x y| = |x| |y| b% considerin& the cases
where x, y < ( and x, y Z ( se$aratel%.
@
Proo )y Contr!&iction?
To $rove that a $ro$osition p is true we ma% assume that p is true and
then show that it would lead to a contradiction or a false statement.
0#am$le? Prove that G2 is irrational.
*olution? /et p? G2 is irrational. Now assumep is true+ that is+ G2 is rational. Then
G2 F a/b which has been reduced+ that is for some inte&ers a and b with no
common factors. 8ence a
2
F 2b
2
which means that a
2
is even and so is a,
sa% a F 2c with inte&er c. *ubstitutin& %ields 7c
2
F 2b
2
or 2c
2
F b
2
hence
b is also even. This means that a and b have a common factor 2 which is a
contradiction+ and so p must be false and p is true.
2.@ Prove the followin& $ro$ositions.
aL The number
3
2 is irrational.
bL The number G2 J G2 is irrational.
cL The number 3 J G2 is irrational.
dL There is no lar&est natural number.
Pro(ing E/i#tence St!tement#?
To $rove a $ro$osition in the form x PKxL+ it suffices+ when $ossible+ to
find one value of x for which PKxL is true.
0#am$le? Prove that there e#ists an irrational number.
*olution? /et PKxL? x is irrational. We will $rove x PKxL b% showin& that PKG2L is
true. This was done is Problem ..
2.( Prove the followin& $ro$ositions.
aL There is a $ositive inte&er n such that n
2
C 2n R F (.
bL There is a real number x such that x
2
C x F 5.
cL There is an inte&er n such that Gn is also an inte&er.
dL There are two real numbers x and y such that x
2
J y
2
F Kx J yL
2
.
eL There is an inte&er n such that n mod 5 F 2 and n mod = F 7.
2. Prove that there are irrational numbers a and b such that a
b
is rational.
Pro(ing 0ni*%ene##?
To $rove a $ro$osition in the form Y x PKxL we first $rove x PKxL and
then $rove the $ro$osition PKx

L PKx
2
L V x

F x
2
.
0#am$le? Prove that there is a uni.ue inte&er x such that 2x J @ F 3.
*olution? /et PKxL? 2x J @ F 3. :irst PKC3L is true K'hec!YL so we $roved x PKxL.
Ne#t su$$ose PKx

L and PKx
2
L are both true. Then 2x

J @ F 3 F 2x
2
J @ V
2x

J @ F 2x
2
J @ V 2x

F 2x
2
V x

F x
2
. 8ence we $roved Y x PKxL.
2.2 Prove the followin& $ro$ositions.
aL There is a uni.ue real number x such that a J x F a for an% number a.
bL There is a uni.ue real number x such that ax F a for all real numbers a.
(
cL /et a be an% inte&er. There is a uni.ue inte&er x such that a J x F (.
dL /et a be an% non"6ero rational number. There is a uni.ue rational
number x such that ax F .
Pro(ing Not1A"" St!tement#?
To $rove the $ro$osition x PKxL it suffices+ when $ossible+ to show that
x PKxL is true. At least intuitivel%+ we ma% see that x PKxL is the
ne&ation of the $ro$osition x PKxL.
2.3 Prove the followin& $ro$ositions.
aL Not all real numbers satisf% x x
2
[ (.
bL Not all real numbers have Kx J yL
2
F x
2
J y
2
.
cL Not for all real numbers+ we have |x J y| = |x| |y|.
dL Not for all natural numbers+ 2
n
> nY
eL Not for all natural numbers+ 3
n
> nY
2.7 Prove that the followin& $ro$osition is false b% showin& that its ne&ation is true?
There is a uni.ue real number x such that 2x
2
C 3x F 2.
Proo )y M!them!tic!" In&%ction?
To $rove a $ro$osition in the form n PKnL where n is a natural number+
it suffices to $rove PKL and PKnL V PKnJL.
0#am$le? Prove the followin& formula for all natural numbers n.
J 3 J 5 J > J @ J ... J K2n C L F n
2
*olution? /et PKnL? J 3 J 5 J > J @ J ... J K2n C L F n
2
We shall $rove n PKnL in two ste$s?
L PKL? F
2
so this $ro$osition is true.
2L PKnL? J 3 J 5 J > J @ J ... J K2n C L F n
2
V J 3 J 5 J > J @ J ... J K2n C L J K2n J L F n
2
J K2n J L
V J 3 J 5 J > J @ J ... J K2n C L J K2n J L F Kn J L
2
V PKnJL
2.5 Prove the followin& formulas for all natural numbers n.
aL J 2 J 3 J 7 J 5 J ... J n F \ n Kn J L
bL 2 J 7 J = J R J ( J ... J 2n F n
2
J n
cL J 2 J 7 J R J = J ... J 2
n
C
F 2
n
C
dL J 3 J @ J 2> J R J ... J 3
n
C
F \ K3
n
C L
eL J 7 J @ J = J 25 J ... J n
2
F nKn J LK2n J L E =
2.= Prove b% induction for all natural numbers n.
aL K2
2n
C L mod 3 F (
bL > divides K2
3n
C L
cL Kn
3
J 2nL mod 3 F (

dL Kn
5
C nL is a multi$le of 5.
eL > divides K2
nJ
J 3
2nC
L
The Princi'"e o M!them!tic!" In&%ction used in the last method of $roof can be
stated b% the $ro$osition PKL ]PKnL V PKnJL^ V n Z PKnL. -ther variations
of this $rinci$le can sometimes be a$$lied. The followin& are some of them.
L In&%ction with )!#e k?
PKkL ]PKnL V PKnJL^ V n Z k PKnL
2L C%m%"!ti(e In&%ction?
PKL ]PKL PK2L ... PKnL V PKnJL^ V n Z PKnL
3L 2o%)"e In&%ction?
m Z PKm+L ]PKm, nL V PKm, nJL^ V m Z n Z PKm, nL
2.> Prove b% induction for the &iven base.
aL n < 2
n
for all n Z
bL 2
n
< nY for all n Z 7
cL 3
n
< nY for all n Z >
dL 2
n
> n
2
for all n Z 5
eL nY < n
n
for all n Z 2
2
'ha$ter 3
Set#
A #et is a collection of ob<ects called the e"ement# of the set. The orderin& of the
elements is not im$ortant and re$etition of elements is i&nored+ for e#am$le ]+ 3+ +
2+ 2+ ^ F ]+ 2+ 3^. A set ma% also be em$t% and it is denoted b% _ or ] ^. )f x is an
element of the set A then we write x A, while the ne&ation is written x A.
*et notations can be ver% convenient. :or e#am$les we ma% redefine the number sets
&iven in 'ha$ter ( as follow. 8ere the notation A F ]x P PKxL^ means that the set A
consists of the elements x for which PKxL is true.
F ](+ I+ I2+ I3+ ...^
F ]x P x > (^
F ]2n P n ^
F ]x P x ^
F ]a/b P a b ^
F ]x P x ^
:or an% two sets A and B, define the followin& set o$erations.
L The %nion A B F ]x P x A x B^
2L The inter#ection A ` B F ]x P x A x B^
3L The &ierence A C B F ]x P x A # B^
7L The #ymmetric &ierence A B F ]x P x A W x B^
:or e#am$les if A F ]+ 2+ 3+ 7+ 5^ and B F ](+ 2+ 7+ =^ then A B F ](+ + 2+ 3+ 7+ 5+
=^+ A ` B F ]2+ 7^+ A C B F ]+ 3+ 5^+ B C A F ](+ =^+ and A B F ](+ + 3+ 5+ =^. Also
we can see that F + F + ` F _+ C F + etc.
3. /et A F ]+ 2+ 3+ 7+ 5^+ B F ](+ 2+ 7+ =^ and C F ]+ 3+ 5^. :ind the followin& sets.
aL KA CL KA ` CL
bL A KB CL
cL KA BL C KA CL
dL KA C BL KA C CL
These set o$erations can be illustrated usin& 3enn &i!gr!m#+
or truth tables+ in which the value is true if x is an element of the set and false if not.
A B A 4 B A B A 5 B A B
T T T T $ $
T $ $ T T T
$ T $ T $ T
$ $ $ $ $ $
3
3.2 True or :alseX 2se ,enn dia&rams or truth tables to verif%.
aL KA BL KA ` BL F A B
bL KA BL KB AL F A B
cL KA BL B F A
dL KA BL B F A
eL A A F A A
4efine the com'"ement of a set A to be A F ]x P x A^. :or e#am$le F .
Theorem? The followin& set identities are the analo& of lo&ical e.uivalences.
L A B F B A
A ` B F B ` A
2L A KB CL F KA BL C
A ` KB ` CL F KA ` BL ` C
3L A KB ` CL F KA BL ` KA CL
A ` KB CL F KA ` BL KA ` CL
7L KAL F A
KA BL F A ` B
KA ` BL F A B
5L A B F A ` B
A B F KA BL KB AL
Two sets are &i#+oint if their intersection is em$t%? A ` B F _. :or e#am$le and
are dis<oint+ and so are and .
3.3 Prove that if A and B are dis<oint then A C B F A and A B F A B.
A set ! is a #%)#et of a set A if x ! V x A. This relation can be written ! A or
sometimes A !. :or e#am$le ]+ 3^ ]+ 2+ 3+ 7^+ + + etc.
3.7 Prove the followin& statements.
aL _ A
bL A A
cL A ` B A
dL A A B
eL A B B C V A C
3.5 Prove that if A B then
aL A B F B
bL A ` B F A
cL A C B F _
dL A B F B A
Theorem? A F B W A B B A
3.= 2se the above theorem to $rove the followin& identities.
aL KA BL = A ` B
bL A B = A ` B
7
The 'ower #et of a set A is defined b% PKAL F ]! P ! A^. 8ence PKAL is the set
consistin& of all the subsets of A.
0#am$le? :ind PKAL for A F ]+ 2^.
*olution? A has a total of four subsets namel% ]^+ ]2^+ _+ and A itself.
8ence PKAL F ]_+ ]^+ ]2^+ A^.
3.> :ind PKAL for each set A.
aL A F ]+ 2+ 3^
bL AF ]+ 2+ 3+ 7^
cL A F _
dL A F PK_L
eL A F PKPK_LL
3.R Prove the followin& statements.
aL PKA ` BL F PKAL ` PKBL
bL PKA BL PKAL PKBL
cL A B W PKAL PKBL
The c!r&in!"ity of a set A is the number of elements in A, denoted b% |A|. :or
e#am$le P]+ 3+ 5+ >^P F 7+ P_P F (+ and PP F O.
Theorem? )f |A| = n then PPKALP F 2
n
K0ver% set with n elements has 2
n
subsets.L
3.@ Prove the above theorem b% cumulative induction.
The cro## 'ro&%ct of A and B is the set A Q B F ]Ka, bL P a A b B^.
0#am$le? )f A F ]+ 2+ 3^ and B F ]x, y^ then
A Q B F ]K+ xL+ K+ yL+ K2+ xL+ K2+ yL+ K3+ xL+ K3+ yL^
B Q A F ]Kx, L+ Kx, 2L+ Kx, 3L+ Ky, L+ Ky, 2L+ Ky, 3L^
Theorem? )f |A| = m and |B| = n then |A"B| = mn.
3.( Prove the followin& statements.
aL A Q B F B Q A W A = B
bL A Q KB Q CL D KA Q BL Q C
cL A " KB ` CL F KA " BL ` KA " CL
dL A " KB CL F KA " BL KA " CL
/et ! be an% set of sets. The gener!"i6e& %nion and gener!"i6e& inter#ection over !
are defined as follow.
L
A!
A = ] x PA! , xA^
2L
A!
A =] xP A! , xA^
:or e#am$le let A
n
be the interval N(+ EnM and ! F ]A
n
P n ^. Then the &enerali6ed
union and intersection over ! are A
n
F N(+ M and `A
n
F ](^.
5
'ha$ter 7
Re"!tion#
A re"!tion on a set A means a subset of A Q A. :or e#am$le if A F]+ 2+ 3^ then the
followin& are some+ but not all+ $ossible relations on A.
L # F ]K+L+ K+2L+ K+3L^
2L # F ]K2+3L^
3L # F ]K+L+ K+2L+ K+3L+ K2+2L+ K2+3L+ K3+3L^
7L # F _
7. )f PAP F n then how man% different relations on A are $ossibleX
)f # is a relation on A then the in(er#e of # is the relation #
C
F ]Kb, aL P Ka, bL #^.
:urthermore if ! is another relation on A then the com'o#ition of # with ! is the
relation ! a # F ]Ka, cL P Ka, bL # K b, cL !^. )n $articular we define #
2
F # a #+
#
3
F #
2
a #+ etc.
0#am$le? /et A F ]+ 2+ 3+ 7^ and # F ]K+2L+ K2+3L+ K2+7L+ K3+3L+ K7+L^ and
! F ]K+3L+ K2+2L+ K3+L+ K3+3L^. Then
#
C
F ]K2+L+ K3+2L+ K7+2L+ K3+3L+ K+7L^
! a # F ]K+2L+ K2+L+ K2+3L+ K3+L+ K3+3L+ K7+3L^
# a ! F ]K+3L+K2+3L+ K2+7L+ K3+2L+ K3+3L^
#
2
F # a # F ]K+3L+ K2+3L+ K2+L+ K3+3L+ K7+2L^
7.2 /et A F ]+ 2+ 3+ 7^ and # F ]K+2L+ K2+L+ K2+7L+ K3+3L+ K7+L+ K7+3L^ A Q A.
aL :ind #
C
and K#
C
L
C
bL :ind #
2
and #
3
cL :ind # a #
C
and #
C
a #
dL :ind K#
C
L
2
and K#
2
L
C
7.3 Prove that # a K# a #L F K# a #L a #. 8ence we ma% write #
3
F # a # a #.
Pro$erties of a relation # A Q A.
L re"e/i(e if a A Ka, aL #
2L #ymmetric if a A b A+ Ka, bL # V Kb, aL #
3L !nti1#ymmetric if a A b A+ Ka, bL # K b, aL # V a = b
7L tr!n#iti(e if a, b, c A+ Ka, bL # K b, cL # V Ka, cL #
0#am$le? /et A F ]+ 2+ 3^ and consider three relations on A?
# F ]K+L+ K+2L+ K2+L+ K2+2L+ K3+3L^
! F ]K+L+ K+3L+ K2+2L+ K3+2L^
$ F ]K+2L+ K+3L+ K2+3L^
# is refle#ive+ s%mmetric+ and transitive+ but not anti"s%mmetric.
! is anti"s%mmetric+ but not refle#ive+ not s%mmetric+ and not transitive.
$ is anti"s%mmetric and transitive+ but not refle#ive and not s%mmetric.
=
7.7 /et A F ]+ 2+ 3+ 7^. Which $ro$erties above are true for each relation # on AX
aL # F ]Ka, bL A Q A P a [ b^
bL # F ]K+L+ K+2L+ K+3L+ K+7L+ K2+2L+ K2+7L+ K3+3L+ K7+7L^
cL # F ]K+L+ K+3L+ K2+L+ K2+2L+ K2+7L^
dL # F ]Ka, bL A Q A P a b > 5^
7.5 /et A F ]+ 2+ 3+ 7^. 5ive an% e#am$le of a relation # on A which is
aL refle#ive+ not anti"s%mmetric+ not transitive.
bL not refle#ive+ not s%mmetric+ not transitive.
cL s%mmetric and transitive.
dL neither s%mmetric nor anti"s%mmetric.
eL both s%mmetric and anti"s%mmetric.
7.= /et # be a relation on A. Prove the followin& $ro$ositions.
aL # is s%mmetric if and onl% if #
C
F #.
bL # is anti"s%mmetric if and onl% if # ` #
C
]K a, aL P a A ^.
cL # is transitive if and onl% if #
2
#.
A relation # A Q A can be re$resented b% a &igr!'h in which each element of A is
re$resented b% a (erte/ and each element Ka, bL # is re$resented b% an e&ge with
direction from a to b. )n the case a = b the ed&e is called a "oo'.
0#am$le? A F ]+ 2+ 3+ 7^ and # F ]K+ 7L+ K2+ L+ K2+ 2L+ K7+ L+ K7+ 2L+ K7+ 3L^.
7.> 4raw the di&ra$h for each of the relations in Problem 7.7.
7.R 8ow can %ou tell from the di&ra$h if # is
aL refle#ive
bL !nti1re"e/i(e Nmeanin& that a A % Ka, aL #M
cL s%mmetric
dL anti"s%mmetric
eL transitive
# A Q A is called an e*%i(!"ence re"!tion if it is refle#ive+ s%mmetric+ and
transitive.
7.@ Prove that the followin& relations are e.uivalence relations.
aL # F ]Ka, bL Q P NaM F NbM^
bL # F ]Ka, bL A Q A P a mod 3 F b mod 3^ where A F ](+ + 2+ ...+ @^
cL # F ]Ka, bL Q P a mod 5 F b mod 5^
dL # F ]Ka, bL A Q A P a = b^ where A F ]+ 2+ 3+ 7^
eL # F ]Ka, bL Q P a b is even^
>
)f # is an e.uivalence relation on A then A is $artitioned into subsets or classes of the
forms Ax F ]a A P Ka, xL #^ for ever% x A. These subsets of A are called the
e*%i(!"ence c"!##e# of A under # and the% satisf% the followin& $ro$erties.
L Kx, yL # V Ax F Ay
2L Kx, yL # V Ax ` Ay F _
3L Ka, bL # W x A, a Ax b Ax
0#am$le? The followin& di&ra$h shows that # is an e.uivalence relation. KWh%XL
There are three e.uivalence classes namel%
A F ]+ 7^ F A7
A2 F ]2^
A3 F ]3+ 5+ =^ F A5 F A=
7.( :ind the e.uivalence classes for each relation in Problem 7.@.
7. 4efine the congr%ence relation on b% a b if and onl% if a mod n F b mod n.
/et # F ]Ka, bL Q P a b^. Prove that # is an e.uivalence relation on and
find the e.uivalence classes.
# A Q A is called a '!rti!" or&er relation if it is refle#ive+ anti"s%mmetric+ and
transitive.
7.2 Prove that the followin& relations are $artial orderin&.
aL A F ]5(+ 22+ 35+ 7^ and # F ]Ka, bL A Q A P a b^
bL A F ]+ 2+ =+ 2+ 27^ and # F ]Ka, bL A Q A P a divides b^
cL A F ]2+ 3+ =+ (+ 2(+ 3(^ and # F ]Ka, bL A Q A P a divides b^
dL # F ]Ka, bL Q P a divides b^
)f # is a $artial order relation then its di&ra$h can be sim$lified into a H!##e &i!gr!m
after these four ste$s?
L 4o not draw loo$s.
2L 4o not draw Ka, cL whenever there are Ka, bL and Kb, cL.
3L Redraw the remainin& &ra$h so that all ed&es $oint u$ward.
7L 4o not draw the directions.
0#am$le? The followin& di&ra$h shows that # is a $artial order relation. KWh%XL
The four ste$s above lead to the 8asse dia&ram of #.
7.3 4raw the 8asse dia&ram for each $artial order relation in Problem 7.2.
A $artial order relation # on A is called a tot!" or&ering if it satisfies one additional
$ro$osition? a A b A+ Ka, bL # Kb, aL #.
R
7.7 Which of the relations &iven in Problem 7.2 are total orderin&X *how that the
8asse dia&ram of a total orderin& can alwa%s be drawn as a strai&ht line.
7.5 Prove that the relation a b &ives a total orderin& on .
*u$$ose # is a $artial order relation on the set A. An element & A is called a "e!#t
e"ement under # if a A+ K&, aL #. Now # is called a we"" or&ering on A if ever%
non"em$t% subset of A has a least element.
7.= Which ones of the sets A &iven in Problem 7.2 have a least element under #X
Which relations are well order relationsX
7.> Prove that a well orderin& is a total orderin& but not conversel%.
7.R 5ive an e#am$le of a total orderin& on a set which is not a well orderin&.
The We"" Or&ering Princi'"e sa%s that is well ordered under the b[b relation.
Theorem? The Well -rderin& Princi$le is e.uivalent to the Princi$le of
Mathematical )nduction.
)f A F ]+ 2+ 3+ ...+ n^ then a relation # A Q A can be re$resented b% a 6ero1one
m!tri/ ' of si6e n " n where K'L
()
F if K(, )L # and K'L
()
F ( if K(, )L #.
0#am$le? *u$$ose A F ]+ 2+ 3^ and # F ]K+L+ K+3L+ K2+L+ K3+2L+ K3+3L^. Then the
6ero"one matri# of # is ' F

(
( (
(
7.@ Re$resent the relations &iven in Problem 7.7 usin& 6ero"one matrices.
7.2( 'onvert these 6ero"one matrices to di&ra$hs.
aL

( (
( (
(
bL

( (
(
( ( (
cL

( ( ( (
( ( (
( ( (
( (
dL

( ( (
( ( (
( ( (
( ( (
The tr!n#iti(e c"o#%re of # A Q A is the smallest transitive relation containin& #.
Theorem? The transitive closure of # is &iven b% # #
2
... #
n
where n F PAP.
7.2 /et A F ]+ 2+ 3+ 7^. 2se this theorem to find the transitive closure of # A Q A.
aL # F ]K+ 2L+ K2+ L+ K2+ 3L+ K3+ 7L^
bL # F ]K+ L+ K+ 2L+ K2+ L+ K7+ 3L^
cL # F ]K+ L+ K+ 7L+ K2+ L+ K2+ 2L+ K3+ 3L+ K7+ 7L^
dL # F ]K+ 7L+ K2+ L+ K2+ 7L+ K3+ 2L+ K3+ 7L+ K7+ 3L^
7.22 :ind the 6ero"one matri# of the transitive closure for each # in Problem 7.2(.
@
'ha$ter 5
$%nction#
A relation from a set A to another set B means a subset of A Q B. A %nction f from A
to B+ denoted b% f ? A V B+ is a relation such that a A YK a, bL f.
0#am$le? /et # ]+ 2+ 3+ 7^ Q ]x, y, *^ defined b%
L # F ]K+xL+ K2+yL+ K3+*L+ K7+xL^
2L # F ]K+xL+ K2+xL+ K3+xL+ K7+yL^
3L # F ]K+*L+ K2+yL+ K3+xL^
7L # F ]K+yL+ K2+xL+ K3+yL+ K3+*L+ K7+xL^
The first two relations are functions but not the last two.
5. *u$$ose # A Q A. 8ow can we tell from the di&ra$h+ or the 6ero"one matri#+
whether or not # is a function from A to AX
5.2 Which ones of the 6ero"one matrices in Problem 7.2( re$resent a functionX
5.3 Which relations are functionsX
aL # F ]Ka, bL Q P b = a J ^
bL # F ]Ka, bL Q P b F NaM^
cL # F ]Ka, bL Q P b
2
F a^
dL # F ]Ka, bL Q P b F Ga^
)f f ? A V B is a function then the statement Ka, bL f can also be written f KaL F b.
The set A in this relation is called the &om!in of f while B the co&om!in of f . The
r!nge of f is the subset of B &iven b% f KAL F ]f KaL B P a A^. )n 'alculus a
function is sometimes &iven in the form y F f KxL whereas its domain and ran&e ma% be
im$licit. :or e#am$le f KxL F x
2
is reall% the function f F ]Kx, x
2
L P x ^ with domain
and ran&e N(+ OL.
5.7 :ind the lar&est $ossible domain and ran&e of each function.
aL f KxL F PxP
bL f KxL F Gx
cL f KxL F Ex
dL f KxL F EGx
eL f KxL F NxM
5.5 /et f ? A V B be a function and let ! and $ be subsets of A. Prove the followin&.
aL f K! $L F f K!L f K$L
bL f K! ` $L f K!L ` f K$L
Pro$erties of a function f ? A V B.
L f is one1to1one or an in+ection if f KaL F f KaAL V a F aA.
2L f is onto or a #%r+ection if f KAL F B.
3L f is a )i+ection if both one"to"one and onto.
0#am$le? All the followin& are functions f ? A V B.
2(
L A F ]+ 2+ 3^+ B F]x, y, *, +^+ f F ]K+yL+ K2+*L+ K3++L^
2L A F ]+ 2+ 3^+ B F]x, y, *, +^+ f F ]K+yL+ K2++L+ K3++L^
3L A F ]+ 2+ 3^+ B F]x, y, *^+ f F ]K+yL+ K2+*L+ K3+xL^
7L A F ]+ 2+ 3+ 7^+ B F]x, y, *^+ f F ]K+yL+ K2+*L+ K3+yL+ K7+xL^
The first is one"to"one but not onto.
The second is neither one"to"one nor onto.
The third is both one"to"one and onto.
The fourth is onto but not one"to"one.
5.= )s f one"to"oneX ontoX bothX
aL f F ]Ka, bL Q P b F a J ^
bL f F ]Ka, bL Q P b F NaM^
cL f F ]Ka, bL Q P b F Ca^
dL f F ]Ka, bL Q P b F 2a^
The in(er#e of a function f ? A V B is the relation f
C
B Q A &iven b% f
C
KbL F a W
f KaL F b. Note that f
C
ma% or ma% not be a function. Moreover if ! B then the
in(er#e im!ge of ! is the subset of A &iven b% f
C
K!L F ]a A P f KaL !^.
5.> :ind f
C
for each function &iven in Problem 5.7. )s f
C
a functionX
5.R Re$eat the .uestion usin& Problem 5.5.
5.@ /et f ? A V B be a function and let ! and $ be subsets of B. Prove the followin&.
aL f
C
K! $L F f
C
K!L f
C
K$L
bL f
C
K! ` $L F f
C
K!L ` f
C
K$L
*u$$ose there are two functions f ? A V B and , ? B V C. The com'o#ition function
,af ? A V C is defined b% ,af KaL F c W f KaL F b ,KbL F c. )n $articular when A F B
F C this definition coincides with that of arbitrar% relations on A.
5.( :ind ,af. Assume %ou !now the a$$ro$riate domain and ran&e for each.
aL f KxL F x+ ,KxL F x
2
bL f KxL F x J + ,KxL F x C
cL f KxL F 2x J + ,KxL F x
2
C 2
dL f KxL F Ex+ ,KxL F Ex
5. *u$$ose f
C
? B V A is a&ain a function. Prove that f
C
af KaL F a a A and
that f af
C
KbL F b b B. ,erif% these facts usin& each function &iven in Problem 5.5
when a$$licable.
Theorem? The inverse of f ? A V B is a&ain a function if and onl% if f is a bi<ection+
in which case f
C
? B V A is also a bi<ection.
2
'ha$ter =
C!r&in!"ity
A set is called inite or ininite de$endin& whether its number of elements is finite or
infinite+ res$ectivel%.
=. *u$$ose both A and B are finite sets. Prove the followin& statements.
aL in<ection f ? A V B W PAP [ PBP
bL sur<ection f ? A V B W PAP Z PBP
cL bi<ection f ? A V B W PAP F PBP
dL )f PAP F PBP then an% function f ? A V B is one"to"one if and onl% if onto.
We now &enerali6ed the definition of cardinalit% to infinite sets. :or arbitrar% set A we
associate to it a c!r&in!" n%m)er PAP satisf%in& the followin& $ro$erties.
L PAP F PBP if bi<ection f ? A V B
2L PAP [ PBP if in<ection f ? A V B
3L PAP < PBP if PAP [ PBP P AP D PBP
Note that the above definitions coincide with the $ro$erties of cardinalit% for finite
sets.
Theorem? PAP [ PBP P BP [ PAP V PAP F PBP K'antor"*chroeder";ernsteinL
4efine PP F o and call a set A co%nt!)"e if PAP [ o or %nco%nt!)"e if PAP > o .
:or e#am$le is itself countable under the bi<ection f KnL F n n .
=.2 Prove that the followin& sets are countable.
aL ` N+(((M
bL `
cL
dL
Theorem? :or an% set A+ e#actl% one of the followin& statements must be true?
L PAP < o
2L PAP F o
3L PAP > o
=.3 Prove that A is finite if and onl% if PAP < o .
The above $roblem sa%s that all finite sets are countable+ but not conversel% since
there e#ist countable sets which are infinite such as . )n some Mathematics boo!s+
an infinite set which is countable is called &en%mer!)"e while in other boo!s the
definition of countable sets does not include finite sets.
=.7 Prove the followin& statements.
aL A subset of a countable set is countable.
bL The union of two countable sets is countable.
22
cL The cross $roduct of two countable sets is countable.
dL The countable union of countable sets is countable.
=.5 Prove that and are both countable. )n $articular PP F PP F o .
Theorem? is uncountable. K'antorL
We define PP F c, the c!r&in!"ity o the contin%%m.
=.= Prove that PAP < PPKALP for an% set A.
Problem =.= im$lies that o F PP < PPKLP and so PKL is also uncountable. )n
$articular it can be shown that PPKLP F c. C!ntor7# Contin%%m Hy'othe#i# asserts
that there is no cardinal number strictl% between o and c. There are however
cardinal numbers lar&er than c+ for instance PPKLP+ PPKPKLLP+ etc.
23

You might also like