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)