0% found this document useful (0 votes)
13 views61 pages

Lenstra Chapter 2

Chapter 2 continues with basic concepts of sets, functions, sequences, and sums, emphasizing the importance of clarity in set notation and operations. It covers definitions of sets, cardinality, operations such as union and intersection, and introduces functions as mappings from one set to another. The chapter concludes with properties of functions and their operations, including composition and various types of increasing and decreasing functions.

Uploaded by

Alejandro Benita
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
13 views61 pages

Lenstra Chapter 2

Chapter 2 continues with basic concepts of sets, functions, sequences, and sums, emphasizing the importance of clarity in set notation and operations. It covers definitions of sets, cardinality, operations such as union and intersection, and introduces functions as mappings from one set to another. The chapter concludes with properties of functions and their operations, including composition and various types of increasing and decreasing functions.

Uploaded by

Alejandro Benita
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 61

Chapter 2, continuation of basic material:

sets, functions, sequences, and sums

here:
1. brief review of basic set-related concepts
2. brief mention of functions
3. focus on sequences and sums

1&2 (sets and functions)


if not thoroughly familiar with this
material, carefully read Chapter 2
using un-axiomatic treatment: a set is an
unordered collection of distinct objects
A is the set of primes less than 13:
A = {2,3,5,7,11}
= {3,7,11,5,2} = {2,2,3,5,5,5,7,11}
2, 5 A: 2 and 5 belong to A, are elements of A
B is the set of non-negative integers at most 100:
B = {0, 1, 2, …, 100}
note usage of “{”, “}” and “…” (ellipses)
be unambiguous: A = {2, 3, …, 11} is inadequate
Set builder notation:
for propositional function P(x)
“S = {x | P(x) }” and “S = {x : P(x) }”
are both short-hands for
x(x S P(x) )
S is the set of all x such that P(x) holds
(in some implicit domain that is often omitted)
examples:
A = {p | p prime and p < 13}
B = {n | n integer and 0 n 100}
D = {n | n = 2m for an integer m} (even integers)
again: always be clear and unambiguous
Common sets
• N is the set of natural numbers
• for some 0 N, others prefer 0 N
no big deal, as long as you’re clear
• B = N 100
• Z is the set of the integers (D = 2Z)
• Q is the set of the rational numbers
• R is the set of the real numbers
• C is the set of the complex numbers
Cardinality of a set S, denoted |S| or #S
|S| = #S = number of distinct elements of S
|A| = #A = 5
|B| = #B = 101
A and B are examples of finite sets
examples of infinite sets:
#N = #Z = #Q =
#R = #C =
and: #N = #Z = #Q #R = #C
empty set, the set without elements: (={})
singleton set, a set with a single element
example: { }, set containing the empty set
equality between sets A and B:
A = B if and only if x (x A x B)
subset: set A is subset of set B
if and only if x (x A x B)
notation: A B (similar: A B x (x B x A))
proper subset: set A is proper subset of set B
if and only if A B and A B
notation: A B (careful with versus )
thm: for every set A: A and A A
(prove A using a vacuous proof)
Power set P(A) of set A: set of all subsets of A
for every set A: A and A A,
thus P(A) and A P(A)
Let A = {1,2,3}, then P(A)=
{ ,{1},{2},{3},{1,2},{1,3},{2,3},{1,2,3}}
note: elements of P(A) are (sub)sets, elements
of these (sub)sets may again be (sub)sets:
Let B = , then P(B) = { }, so P( ) = { }
Let C = P( ) = { },
P(C) = { , { }}, so P(P( )) = { , { }}
Let D = P(P( )) = { , { }}, so P(D) =
P(P(P( ))) = { ,{ },{{ }},{ ,{ }}}
fye, power set of P(P(P( ))) = { ,{ },{{ }},{ ,{ }}}:
P(P(P(P( )))) = {
,
{ }, {{ }}, {{{ }}}, {{ ,{ }}},
{ ,{ }}, { ,{{ }}}, { ,{ ,{ }}},
{{ },{{ }}}, {{ },{ ,{ }}},
{{{ }},{ ,{ }}},
{ ,{ },{{ }}}, { ,{ },{ ,{ }}},
{ ,{{ }},{ ,{ }}}, {{ },{{ }},{ ,{ }}},
{ ,{ },{{ }},{ ,{ }}}
}
Cartesian product A B of sets A and B:
set of all ordered pairs (a,b), a A and b B:
A B = {(a,b) | a A b B}
example:
A = {H,L,S}, B = {A,B,L’,P}:
A B = { (H,A),(H,B),(H,L’),(H,P),
(L,A),(L,B),(L,L’),(L,P),
(S,A),(S,B),(S,L’),(S,P) }
#(A B) = #A # B = 3 4 = 12,
relation from A to B: a subset of A B
example:
{(H,B),(H,L’),(H,P),(L,P)} A B
for sets A, B, C
A B C = {(a,b,c) | a A b B c C}
but
(A B) C = {(d,c) | d A B c C}
Set operations
to create new sets from existing sets (similar
to using logical operators to create compound
propositions from existing propositions)
complement: A = {x| x A} = {x| (x A)}
(always with respect to some universe U)
union: A B = {x| x A x B}
intersection: A B = {x| x A x B}
(A and B disjoint if A B = )
difference: A B = A\B = {x| x A (x B)}
symmetric difference:
A B = A B = {x| x A x B}
Note: correspondence with logical operations (and “ ” “ ”)
set operations lead to set identities (page 132 (124))
such as
A (B C) = (A B) (A C) (distributive law)
A B A B
(De Morgan’s laws)
A B A B

which can proved
1. with membership tables
2. using both and
3. “directly”
example: Prove A B A B
1. with membership table (i.e., truth table for x A, etc.):

A B A B A B A B A B
0 0 0 1 1 1 1
0 1 1 0 1 0 0
1 0 1 0 0 1 0
1 1 1 0 0 0 0

2. using both and , thus proving:


A B A B and A B A B
3. directly
using and to prove that A B A B
: let x A B
x A B
( x A B)
( x A x B)
( x A) ( x B)
( x A) ( x B)
x A x B
x A B
it follows that A B A B
all “ ” can be replaced by “ ” (or “ ”),
from which “ ” follows as well
direct proof of A B A B

A B {x | x A B}
{x | ( x ( A B))}
{x | ( x A x B)}
{x | ( x A) ( x B)}
{x | ( x A) ( x B)}
{x | x A x B}
{x | x A B}
A B
prove A (A B)=A by showing and
• A (A B) A:
if x A (A B), then x A or x A B,
so:
x A
or
(x A and x B)
in either case x A
it thus follows that A (A B) A
• A (A B) A:
if x A, then x A (A B)
it thus follows that A (A B) A
Note on Venn diagrams
• Venn diagrams are pictures of sets,
drawn as subsets of some universal set U
• may be used for pictorial purposes
but never for proofs
• three sets intersecting
in all possible ways:

• four sets:
Note on Venn diagrams
• Venn diagrams are pictures of sets,
drawn as subsets of some universal set U
• may be used for pictorial purposes
but never for proofs
• 5, 7, and 11 sets intersecting
in all possible ways:
Returning to sets, a note on cardinalities
given finite sets A and B, what is |A B| ?
|A| is the cardinality of A
|B| is the cardinality of B
|A| + |B| is the cardinality of the union
A B of A and B, where all elements that
belong to both A and B are counted twice
thus: |A| + |B| = | A B | + | A B |
equivalently: | A B | = |A| + |B| | A B |
known as
the principle of inclusion and exclusion
(and an example of “proof by intimidation”; how to really prove this? )
Inclusion/exclusion example
A = {n Z : 0 n 100, n multiple of 5}
= {n Z : 0 n 100, 5|n}
B = {n Z : 0 n 100, 7|n}
|A| = 21, |B | = 15
what is |A B| ?
A B = {n Z:0 n 100, 5|n or 7|n }
|A| + |B| = 21 +15 = 36
counts multiples of both 5 and 7 twice:
A B = {n Z : 0 n 100, 5|n and 7|n }
= {0, 35, 70}
|A B| = |A| + |B| |A B| = 21 + 15 3 = 33
more complicated
A = {n Z : 0 n 100, 5|n}, |A| = 21
B = {n Z : 0 n 100, 7|n}, |B| = 15
C = {n Z : 0 n 100, 3|n}, |C| = 34
what is |A B C| ?
|A B C|=|A|+|B|+|C| |A B| |A C| |B C|+|A B C|

Proof: Let D = B C, then


|A B C| = |A D|
= |A|+|D| |A D|
= |A|+|B C| |A (B C)|
= |A|+|B|+|C| |B C| |(A B) (A C)|
The result now follows from
|(A B) (A C)| = |A B|+|A C| |(A B) (A C)|
= |A B|+|A C| |A B C|
more complicated
A = {n Z : 0 n 100, 5|n}, |A| = 21
B = {n Z : 0 n 100, 7|n}, |B| = 15
C = {n Z : 0 n 100, 3|n}, |C| = 34
what is |A B C| ?
A B = {0, 35, 70}: |A B| = 3
A C = {n Z : 0 n 100, 3|n and 5|n}
= {0, 15, 30, 45, 60, 75, 90}: |A C| = 7
B C = {n Z : 0 n 100, 3|n and 7|n}
= {0, 21, 42, 63, 84}: |B C| = 5
A B C = {0}: |A B C| = 1
|A B C|=|A|+|B|+|C| |A B| |A C| |B C|+|A B C|
= 21+15+34 3 7 5+1 = 56
Questions?
Concludes 2nd section of Chapter 2
Functions
given nonempty sets A and B,
a function f from A to B is an assignment of
exactly one element of B to each element of A

What does that mean? Can’t we do better?


Functions
first an unusually complicated definition
reminder: a relation from A to B is
an arbitrary subset of A B
A and B nonempty sets, function f from A to B is:
a relation from A to B
such that a A !b B (a,b) f
thus, for each element of A there is
exactly one ordered pair in f whose
first element equals that element of A
note: no limitation on number of pairs in f
in which any b B may appear
Functions, more traditionally:
given nonempty sets A and B,
a function f from A to B is an assignment of
exactly one element of B to each element of A
we say that f maps A to B and write:
• f(a) = b (or (a,b) f as on previous slide):
b is the image of a
a is a preimage of b
• for any element of B, there may be
any number of elements of A mapping to it
function f from A to B
• f: A B
(note: same arrow as before, different meaning)
• f goes from domain A to codomain B
• f has range f(A) = {b B| a A f(a)=b} B
b f(A) a A f(a)=b,
a property that does not necessarily hold for B
• for S A, the image of S under f is defined as
f(S) = {b | b B and s S f(s)=b}
= {f(s) | s S} f(A)
Operations on functions
• sum and product of two functions f, g: A R:
sum: f+g: A R: (f+g)(x) = f(x)+g(x)
product: fg: A R: (fg)(x) = f(x)g(x)
• in general: f, g: A B inherit operations on B
• composition of f: A B and g: B C :
g f :A C : ( g f )( x) g ( f ( x))
Example
f: set of students R3, g: R3 {1, 1.5, 2, 2.5, …, 5, 5.5, 6}
f(Amy) = (H, M, F) is triple of Amy’s average homework grade (H),
midterm grade (M), and final grade (F)
g(x,y,z) = [[0.3x+0.2y+0.5z]] (with [[.]] rounding to nearest half point)
then ( g f )(Amy) is Amy’s overall grade
but ( f g )(Anna ) is not defined
Simple properties of functions
f: A R
• f is increasing:
x A y A x>y f(x) f(y)
• f is strictly increasing:
x A y A x>y f(x) > f(y)
• f is decreasing:
x A y A x>y f(x) f(y)
• f is strictly decreasing:
x A y A x>y f(x) < f(y)
Interesting properties of functions, f: A B
• f is one-to-one or injective or an injection
iff a1,a2 A f(a1) = f(a2) a1= a2
iff a1,a2 A a1 a2 f(a1) f(a2):
no “collisions”
• f is onto or surjective or a surjection
iff f(A) = B
iff b B a A f(a) = b:
everything in B is reached
• f is one-to-one correspondence or bijection
iff f is one-to-one and onto
iff b B !a A f(a) = b
• injection f: A B is bijection f: A f(A)
inverse of a function
injection f: A B, thus bijection f: A f(A)
b f(A) !a A f(a) = b
let g = {(b,a): b f(A), a A, f(a)= b} f(A) A
then g is relation f(A) A such that
b f(A) !a A (b,a) g (i.e., g(b) = a)
where (b,a) g f(a) = b
thus g is a function from f(A) to A such that
g(b) = a if and only if f(a) = b
this g is called the inverse f 1 of f:
function f 1: f(A) A such that
f 1(b) = a if and only if f(a) = b
remarks on inverse
injection f: A B, bijection f: A f(A),
the latter’s inverse f 1: f(A) A
with f 1(b) = a if and only if f(a) = b
• a A f 1(f(a)) = a
1
f f :A f ( A) A , the identity on A
• b f(A) f(f 1(b)) = b
1
f f : f ( A) A f ( A) , identity on f(A)
• it may be the case that f can be computed while
computing f 1 is intractable, or vice versa
examples
2
f: R R defined by f(x) =x2 (f :x x ):
• f not injective: f(1) = f ( 1) = 1
• “same” f: R 0 R is injective
• “same” f: R 0 R is injective too
• f not surjective: y R x R f(x) y (y < 0)
( y R x R f(x) = y )
• “same” f: R R 0 is surjective
• “same” f: R 0 R 0 is bijection
with inverse f 1: R 0 R 0: f 1(y) = y
• or “same” f: R 0 R 0is bijection
with inverse f 1: R 0 R 0 : f 1(y) = y
more examples
• g: R R, g(x) = x2k+1 for k N (g : x x 2k 1 ):
g is injective and surjective, and thus bijective
example of simple non-trivial bijective
correspondence between R and R
• h: R { /2+k : k Z} R, h(x) = tan(x)
h surjective, not injective: k Z h(k )=0
“same” h: ( /2, /2) R ( open interval notation! )
is injective while staying surjective:
h: ( /2, /2) R, h(x) = tan(x), is bijection
implies bijection between ( /2, /2) and R
arctan = tan 1 is bijection R ( /2, /2)
More on cardinalities
sets A and B have by definition the
same cardinality if there is a
bijection between A and B
a set S is countable if S is finite or has
the same cardinality as N
if S countable and infinite: |S| = 0: “aleph null”
countability of S implies that S can be
“enumerated”: S is finite, or if not
there exists a bijection f: N S,
S = {f(i) : i N} = {f(0), f(1), f(2), … }
a set that is not countable is uncountable:
any enumeration will miss (infinitely many) elements
N, Z, Q are countable
to prove this, establish bijections between
• N and N:
the identity map
• Z and N:
define f : Z N:
stretch all “non-negatives” to “even”:
if z 0 then f(z) = 2z
fill the odd holes with the negatives:
if z < 0 then f(z) = (2z + 1)
this f is “obviously” a bijection
with f 1 : N Z, n ( 1)n[(n+1)/2]
• Q and N: next slide
More on cardinalities
sets A and B have by definition the
same cardinality if there is a
bijection between A and B
a set S is countable if S is finite or has
the same cardinality as N
if S countable and infinite: |S| = 0: “aleph null”
countability of S implies that S can be
“enumerated”: S is finite, or if not
there exists a bijection f: N S,
S = {f(i) : i N} = {f(0), f(1), f(2), … }
a set that is not countable is uncountable:
any enumeration will miss (infinitely many) elements
N, Z, Q are countable
to prove this, establish bijections between
• N and N:
the identity map
• Z and N:
define f : Z N:
stretch all “non-negatives” to “even”:
if z 0 then f(z) = 2z
fill the odd holes with the negatives:
if z < 0 then f(z) = (2z + 1)
this f is “obviously” a bijection
with f 1 : N Z, n ( 1)n[(n+1)/2]
• Q and N: next slide
Q is countable – less hand-waving
surjection N 0 Q>0 suffices (hold breath at duplicate)
Let Ik = {(k 1)k/2,1+(k 1)k/2,…,k(k+1)/2 1}
for k = 1, 2, 3, …
then |Ik| = k(k+1)/2 1 (k 1)k/2 +1 = k
I1 = {0}, I2 = {1,2}, I3 = {3,4,5}, I4 = {6,7,8,9}, …
I
k 1 k
N 0 and k Ik I
n N 0 !k n Ik; denote this k by k(n) (=[(1+ (1+8n))/2])

(k(0)=1, k(1)=k(2)=2, k(3)=k(4)=k(5)=3, k(6)=k(7)=k(8)=k(9)=4)


define i(n) = n (k(n) 1)k(n)/2: 0 i(n) < k(n)
k ( n) i ( n)
g : N 0 Q>0 n is surjective
i ( n) 1
R is uncountable – not too precisely
Proof by contradiction: assume R is countable,
implying countability of R1= {x R: 0 < x < 1}
bijection h : N>0 R1 :
h(1) = x1, h(2) = x2, …, h(i) = xi, …
and {x1, x2, …, xi, … } = R1
xi = 0.di1di2di3…dii… is xi’s decimal expansion
for i = 1, 2, 3, …, let i dii, i {0,1,…,9}
(“Cantor diagonalization argument”)
and let y = 0. 1 2 3… i…
y R1 and i y xi
contradiction with {x1, x2, …, xi, …}= R1
(un)countability examples
• the set of real numbers with decimal representation consisting
of just digits “7” and possibly a single decimal point:
7, 77, 7.7, 777, 77.7, 7.77, 7777, 777.7, 77.77, 7.777, …
first list the single one consisting of a single digit, then the two
consisting of two digits, followed by the three consisting of
three digits, etc. countable
• as above, but allow digits 8 as well: use Cantor’s
diagonalization to show that for any enumeration an element
can be found that will not be enumerated by picking 7 if dii=8
and 8 if dii=7 (see previous slide) uncountable
• the set of all finite length bit strings:
0,1, 00,01,10,11, 000,001,010,011,100,101,110,111, …
for k =1, 2, 3, … in succession list the 2k bit strings of length k
(by counting in binary from 0 to 2k 1 and using leading
zeros) countable
Special functions
• rounding:
R Z, x x , the integer nearest to x
(halves rounded down; x goes up)
• floor:
R Z, x x , the largest integer x
• ceiling:
R Z, x x , the smallest integer x
• entier:
R 0 Z, x [x], the integer part of x
• factorial: n
N Z, n n!, with n! i ; note that 0!=1
i 1
example
3x = x + x+1/3 + x+2/3
Proof. let x = n + , with n Z and 0 <1
case analysis:
• if 0 < 1/3, then 3x = 3n + , 0 < 1,
3x = 3n and x = x+1/3 = x+2/3 = n
• if 1/3 < 2/3, then 3x = 3n+1+ , 0 < 1,
3x = 3n+1 and x = x+1/3 = n,
but x+2/3 = n+1
• if 2/3 < 1, then 3x = 3n+2+ , 0 < 1,
3x = 3n+2 and x = n,
but x+1/3 = x+2/3 = n+1
Another example
2x = x + x ½
normally, one takes x = m , with 0 <1
instead, let x = n + , with n Z and 0 < 1,
then x = n + 1
• if 0 < ½, then 2x = 2n + 2 with 0 < 2 1,
so 2x = 2n + 1;
x ½ = n then implies 2x = x + x ½
• if ½ < 1, then 2x = 2n + 2 with 1 < 2 2,
so 2x = 2n + 2;
x ½ = n+1 then implies 2x = x + x ½
Any questions?
Concludes 3rd section of Chapter 2
Introduction to sequences and summations

informally:
a sequence is a possibly infinite ordered list with
a first, a second, a third, a fourth, … element
slightly more formally:
a sequence is a function f from a subset of
the set of natural numbers (with or without 0)
to some other set S:
a1, a2, a3, … S
or
a0, a1, a2, … S
where ai = f(i)
common sequences
• 0, 1, 2, 3, 4, …
sequence of natural numbers, ni = i, i 0
• 0, 2, 4, 6, 8, …
sequence of even numbers 0, mi = 2i, i 0
• 1, 1, 2, 6, 24, 120, 720,…
sequence of factorials, fi = i!, i 0
• 2, 3, 5, 7, 11, 13, 17, 19, …
sequence of primes, pi is ith prime, i 1
• 0, 1, 1, 2, 3, 5, 8, 13, 21,…
Fibonacci sequence:
Fi = i for i = 0, 1, Fi = Fi 2+Fi 1 for i 2
crazy sequences
• 2, 2, 3, 3, 4, 4, 5, 5, 5, 5, 5, 6, …
bi = bitlength of pi, i 1
• 4, 3, 3, 5, 4, 4, 3, 5, 5, 4, 3, 6, …
(in French: 4, 2, 4, 5, 6, 4, 3, 4, 4, 4, 3, 4, …)
• 5, 6, 5, 6, 5, 5, 7, 6, 5, 5, 8, 7, …
(in French: 7, 8, 9, 9, 9, 7, 8, 8, 8, 7, 7, 8, ...)
• given an integer sequence
(such as 171, 277, 367, 561, 567, 18881,…),
how to find what it is?
encyclopedia of integer sequences
http://oeis.org/
Remarks on sequences
sequences do not necessarily consist of integers:
• xi =1/i (i>0)
• yi = ri for r R
sequences are not necessarily infinite:
• si = ith SD student (lexicographically or sciper-wise)
sequences are not necessarily well understood
2i
• 3, 5, 17, 257, 65537, …, primes 2 1
(are there more than five Fermat primes?)
• 3, 5, 7, 11, 13, 17, 19, 29, 31, 41, 43, …
(are there infinitely many twin primes?)
• primes 123456789101112131415…: any?
Common sequences
arithmetic progression: a sequence of the form
a, a+d, a+2d, a+3d,… , a+kd, … for a, d R
with initial term a and common difference d:
ith term ai equals a+id ( i>0 ai ai 1 = d )

geometric progression: a sequence of the form


g, gr, gr2, gr3,… , grk, … for g, r R
with initial term g and common ratio r.
ith term gi equals gri ( i>0 gi/gi 1 = r )
Often needed: summations of sequences
• sum of elements of arithmetic progression
a, a+d, a+2d, a+3d,… , a+kd
• sum of elements of geometric progression
g, gr, gr2, gr3,… , grk,
• and sums of elements of similar sequences
k
for ai = a+id determine a0 a1 ... ak ai
i 0
k
for gi = gri determine g 0 g1 ... g k gi
i 0
need to be familiar with methods
to calculate such sums
Sum of an arithmetic progression
k
ai a id , then a0 a1 a2 ... ak ai
i 0
k k k
(a id ) a id
i 0 i 0 i 0
k
(k 1)a d i
i 0
k (k 1) dk
here we use : (k 1)a d (k 1)(a )
k k k
2 2
i i i /2
i 1 i 1 i 1
let j=k+1 i, thus i=k+1 j; j=k when i=1 and j=1 when i=k; thus
k k k
i i (k 1 j ) / 2
i 1 i 1 j 1
k k k
j (k 1 j ) / 2 ( j (k 1 j )) / 2
j 1 j 1 j 1
k
k (k 1)
(k 1) / 2
j 1 2
Often needed: summations of sequences
• sum of elements of arithmetic progression
a, a+d, a+2d, a+3d, …, a+kd: k
for ai = a+id determine a0 a1 ... ak ai
i 0
• sum of elements of geometric progression
g, gr, gr2, gr3, …, grm: m
for gj = grj determine g 0 g1 ... g m gj
j 0
• sums of elements of related progression
r, 2r2, 3r3, 4r4, …, nrn: n
for t r determine t1 t 2 ... t n t
1
need to be familiar with those sums
and with the methods to calculate them
Sum of a geometric progression, I
k k k
i i
gi gr , then g 0 g1 g 2 ... g k gi gr g ri
i 0 i 0 i 0
k
let S ri; if r 0 then S 1
i 0
k k k
i 1 i 1 1
assume r 0, then S r r , thus S / r r 1/ r ri
i 0 i 0 i 1
let i 1 j , then j 0 if i 1, and j k 1 if i k , thus
k 1 k
j
S / r 1/ r r 1/ r rj rk 1/ r S r k
j 0 j 0

with r 0 it follows that S 1 rS r k 1 and thus, if r 1, that


rk 1 1
S (also valid for r 0; if r 1, then S k 1)
r 1
i 1
note : for 0 r 1 it follows that r
i 0 1 r
Sum of a geometric progression, II
k
another way to compute S ri
i 0

let f ( X ) 1 X X 2 ... X k (then f ( r ) S)


Xf ( X ) X X 2 ... X k Xk 1

k 1 Xk 1 1
thus Xf ( X ) f (X ) X 1 and f ( X ) (if X 1)
X 1
k
cleaner (without dots) : f ( X ) X i , then
i 0
k k k
i i
( X 1) f ( X ) ( X 1) X X X Xi
i 0 i 0 i 0
k k k 1 k k 1 k
i 1 i j i i
X X X X X Xi
i 0 i 0 j 1 i 0 i 1 i 0
k k
k 1 i 0
X X X Xi Xk 1
1
i 1 i 1
Sum of an arithmetic progression
k
ai a id , then a0 a1 a2 ... ak ai
i 0
k k k
(a id ) a id
i 0 i 0 i 0
k
(k 1)a d i
i 0
k (k 1) dk
here we use : (k 1)a d (k 1)(a )
k k k
2 2
i i i /2
i 1 i 1 i 1
let j=k+1 i, thus i=k+1 j; j=k when i=1 and j=1 when i=k; thus
k k k
i i (k 1 j ) / 2
i 1 i 1 j 1
k k k
j (k 1 j ) / 2 ( j (k 1 j )) / 2
j 1 j 1 j 1
k
k (k 1)
(k 1) / 2
j 1 2
k
Similar sum T (r ) ir i 1 , determined in two ways (for r 1)
i 0
k
rk 1 1
1 differentiating S (r ) ri leads to T (r ) S ' (r ) :
i 0 r 1
(k 1)r k (r 1) (r k 1
1) kr k 1
(k 1)r k 1
T (r ) S ' (r )
(r 1) 2 (r 1) 2
2 directly :
k k k
T (r ) ir i 1
ri 1
(i 1)r i 1

i 1 i 1 i 1
k 1 k k 1 k 1
ri r (i 1)r i 2
ri r ir i 1

i 0 i 1 i 0 i 0
k
r 1
r (T (r ) kr k 1 ) T (r ) follows
r 1

(page 166/157 : more summations, will be proved later)


Section 2.6/3.8: matrices
• if you’re not familiar with matrices: read it
• k m rectangles of numbers:
k rows, m columns
• originally to represent linear
transformations from Rm to Rk
• wide variety of applications
Matrix product, traditional computation
k, m, n Z>0:
k m matrix A (aij )ik ,1, mj 1 ,
m n matrix B (b j ) mj ,1, n 1 ,
AB C is k n matrix C (ci ) ik ,1, n 1
m
with ci j 1
aij b j :
ci is inner product of A' s ith row and B ' s th column
• computation in k m n multiplications
(disregarding additions)
• not commutative: even if AB and BA both
defined, they are not necessarily equal
fye, matrix multiplication exponent
• traditional: n n matrices A and B,
computation of AB in n3 multiplications
• can it be done faster?

yes, but no one


knows how fast:

n2.3727 best so far


(compare to integer
multiplication...)

(picture shamelessly copied from wikipedia)

You might also like