Master Theorem

Master Theorem

 Master Theorem is the most useful and easy method to compute the time complexity
function of recurrence relations.

 Recursive functions might get complex if we start calculating its time complexity function
by other commonly used simpler methods.

 Master Theorem is the best method to quickly find the algorithm's time complexity from
its recurrence relation.

The approach was first presented by John Bentley, Dorothea, and James B. Saxe in 1980
Master Theorem

 Master theorem provides an asymptotic analysis for recurrence relations of types that
occur in the analysis of many divide and conquer algorithms.

 The master theorem always yields asymptotically tight bounds to recurrences from divide
and conquer algorithms that partition an input into smaller subproblems of equal sizes
Master Theorem

Master theorem can be applied to:

 Decreasing Functions - Decrease and Conquer Algorithms

 Dividing Functions - Divide and Conquer Algorithms

Master Theorem
(Decreasing Function)
Master Theorem for Decreasing Functions : Decrease and Conquer
We have already seen Master Theorem
T(n) ¿ 𝑇 (𝑛 − 1) + 1 T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
T(n) ¿ 𝑇 (𝑛 − 1) + n a > 0, b > and f(n) = O(n)
where d ≥ 0
T(n) ¿ 𝑇 (𝑛 − 1) + log n
Case 1:
T(n) ¿ 𝑇 (𝑛 −2) + 1
if a =1 O(n * f(n)) or O(nd+1)
T(n) ¿ 2 𝑇 (𝑛 −1) + 1
Case 2:
T(n) ¿ 2 𝑇 (𝑛 −2) + 1 if a >1 O(f(n) a) or O(nd a)
T(n) ¿ 3 𝑇 (𝑛 −1) + 1 Case 3:
T(n)¿ 2 𝑇 (𝑛 − 1)+ n if a <1 O(f(n)) or O(nd)
Master Theorem for Decreasing Functions Case 1
Example 1:
Master Theorem
T(n)¿ 𝑇 (𝑛 − 1)+ 1
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=1 f(n) = 1
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems
d=p=0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
= O( nd+1 n)
Case 1: if a =1 then O(nd+1 logp n ) = O( n )
or O(f(n)*an/b) = O( n 0+1
n)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or
O( n * f(n) ) f(n) = 1
Case 3: if a <1 then O(n log )d p

= O( n )
Master Theorem for Decreasing Functions Case 1
Example 2:
Master Theorem
T(n)¿ 𝑇 (𝑛 − 1)+ n
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=1 f(n) = n
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems
d = 1, p = 0 = n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd+1 n)
= O( n2 )
or O(f(n)*an/b) = O( n1+1 n)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( n * f(n) ) f(n) = n
Case 3: if a <1 then O(nd logp)
= O( n2 )
Master Theorem for Decreasing Functions Case 1
Example 3:
Master Theorem
T(n)¿ 𝑇 (𝑛 − 1)+ log n
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=1 f(n) = log n
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d=0 , p=1 =nn
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd+1 n)
= O( nlog n
or O(f(n)*a ) n/b
= O( n 0+1
n)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( n * f(n) ) f(n) = log n
Case 3: if a <1 then O(nd logp)
= O(n log n)
Master Theorem for Decreasing Functions Case 1
Example 4:
Master Theorem
T(n)¿ 𝑇 (𝑛 − 2)+ 1
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=1 f(n) = 1
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d=0 , p=0 =nn
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd+1 n)
= O( n )
or O(f(n)*a ) n/b
= O( n 0+1
n)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( n * f(n) ) f(n) = 1
Case 3: if a <1 then O(nd logp)
= O(n)
Master Theorem for Decreasing Functions Case 2
Example 5:
Master Theorem
T(n)¿ 𝟐 𝑇 (𝑛 − 1)+ 1
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=2 , b=1 f(n) = 1
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d =0 , p =0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd n an/b)
= O( 2n )
or O(f(n)*an/b) = O( n0 n 2n/1)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( f(n) *an/b ) f(n) = 1
Case 3: if a <1 then O(nd logp)
= O(2n)
Master Theorem for Decreasing Functions Case 2
Example 6:
Master Theorem
T(n)¿ 𝟐 𝑇 (𝑛 −2)+ 1
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=2 , b=2 f(n) = 1
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d =0 , p =0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd n an/b)
= O( 2n/2 )
or O(f(n)*an/b) = O( n0 n 2n/2)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( f(n) *an/b ) f(n) = 1
Case 3: if a <1 then O(nd logp)
= O(2n/2)
Master Theorem for Decreasing Functions Case 2
Example 7:
Master Theorem
T(n)¿ 𝟑 𝑇 (𝑛 − 1)+ 1
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=3 , b=1 f(n) = 1
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d =0 , p =0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd n an/b)
= O( 3n )
or O(f(n)*an/b) = O( n0 n 3n/1)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( f(n) *an/b ) f(n) = 1
Case 3: if a <1 then O(nd logp)
= O(3n)
Master Theorem for Decreasing Functions Case 2
Example 8:
Master Theorem
T(n)¿ 𝟐 𝑇 (𝑛 − 1)+ n
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=2 , b=1 f(n) = n
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d =1 , p =0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd n an/b)
= O( n 2n )
or O(f(n)*an/b) = O( n1 n 2n/1)
Case 2: if a >1 then O(nd logp n a)
or O(f(n)) or O( f(n) *an/b ) f(n) = n
Case 3: if a <1 then O(nd logp)
= O(n 2n)
Master Theorem for Decreasing Functions Case 3
Example 8:
Master Theorem
T(n)¿ 𝟎 . 𝟓 𝑇 ( 𝑛− 1)+ n
T(n) ¿ 𝒂 𝑇 (𝑛− 𝒃) + f(n)
a=0.5 f(n) = n
a > 0, b > 0 , f(n) = O(n) where d, p ≥ 0 =nn
a = number of subproblems d =1 , p =0 =n n
b = the constant by which the subproblem size is reduced.
d = degree of polynomial
or O(n* f(n))
Case 1: if a =1 then O(n d+1
log n )
p = O( nd n)
= O( n )
or O(f(n)*a ) n/b
= O( n n)
1

Case 2: if a >1 then O(nd logp n a)

or O(f(n)) or O( f(n) ) f(n) = n
Case 3: if a <1 then O(nd logp)
= O(n)
Master Theorem
(Dividing Function)
Master Theorem for Dividing Functions : Divide and Conquer

𝒏
T(n)¿ 𝒂 𝑻 ( )+ f(n) T(n) = 𝜽 ( 𝟏)
𝒃

a≥1 : a is the number of subproblems

b>1 : b the factor by which the subproblem size is reduced in each recursive call
n : n is the input size of the problem
𝒏
𝒃 : size of each subproblem (assuming size of each subproblem is same)
: f(n) denotes cost of the work done outside the
f(n) = O(n) where d ≥ 1, p ≥ 1 recursive call, which includes:
d = degree of polynomia  cost of dividing the problem
f(n) denotes amount of time taken at top level of recurrence  cost of merging the solutions
Master Theorem for Dividing Functions : Divide and Conquer

𝒏
T(n)¿ 𝒂 𝑻 ( )+ f(n)
𝒃
where a ≥ 1, b > 1 and

f(n) = O(n) , where d ≥ 1, p ≥ 1

T(n) = 𝜽(𝟏) n is less than some bound k > 0,

the smallest input size that will lead to a recursive call
Master Theorem for Dividing Functions : Divide and Conquer
a < bd
𝒏 logb a < logb bd
T(n)¿ 𝒂 𝑻 ( )+ f(n) T(n) = 𝜽 ( 𝟏) logb a < d logb b
𝒃
logb a < d
where a ≥ 1, b > 1 and f(n) = O(n) , where d ≥ 1, p ≥ 1
logb a = log a / log b

Case 1: If a > bd , (or logb a > d) Case 2: If a = bd , (or logb a = d)

then T(n) = Θ()  If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

Case 3: If a < bd , (or logb a < d)
 If p < -1, then T(n) = Θ ( )
 If p ≥ 0, then T(n) = Θ(nd logp n)

 If p < 0, then T(n) = Θ(nd)

Master Theorem for Dividing Functions : Divide and Conquer

Example 1: Here a = 3 , b = 2
𝒏
T(n)¿ 𝟑 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
3 21 log2 3 1
Case 2: If a = bd , (or logb a = d) 3>2 log2 21.58 1
a > bd 1.58 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 1 1.58 > 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 1: Here a = 3 , b = 2
𝒏
T(n)¿ 𝟑 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
3 21 log2 3 1
Case 2: If a = bd , (or logb a = d) 3>2 log2 21.58 1
a > bd 1.58 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 1 1.58 > 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 1: Here a = 3 , b = 2
𝒏
T(n)¿ 𝟑 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)

then T(n) = Θ() T(n) = Θ()

= Θ()

= Θ()
Master Theorem for Dividing Functions : Divide and Conquer

Example 1b: Here a = 8 , b = 2

𝒏
T(n)¿ 𝟖 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
8 22 log2 8 2
Case 2: If a = bd , (or logb a = d) 8>4 log2 23 2
a > bd 3 log2 2 2
Case 3: If a < bd , (or logb a < d) Case 1 3 > 2
Master Theorem for Dividing Functions : Divide and Conquer

Example 1b: Here a = 8 , b = 2

𝒏
T(n)¿ 𝟖 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
8 22 log2 8 2
Case 2: If a = bd , (or logb a = d) 8>4 log2 23 2
a > bd 3 log2 2 2
Case 3: If a < bd , (or logb a < d) Case 1 3 > 2
Master Theorem for Dividing Functions : Divide and Conquer

Example 1b: a=8,b=2

𝒏
T(n)¿ 𝟖 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 1: If a > bd , (or logb a > d)

8 = 23
then T(n) = Θ()
log2 8 = log2 23
T(n) = Θ() = Θ() Required Answer log2 8 = 3 log2 2
log2 8 = 3
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ 1 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = 1 = n0
Give us d=0,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
1 20 log2 1 0
Case 2: If a = bd , (or logb a = d) 1=1 log2 20 0
a = bd 0 log2 2 0
Case 3: If a < bd , (or logb a < d) Case 2 0 = 0
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ 1 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = 1 = n0
Give us d=0,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
1 20 log2 1 0
Case 2: If a = bd , (or logb a = d) 1=1 log2 20 0
a = bd 0 log2 2 0
Case 3: If a < bd , (or logb a < d) Case 2 0 = 0
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ 1 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = 1 = n0
Give us d=0,p=0
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

 If p < -1, then T(n) = Θ ( )

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ 1 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = 1 = n0
Give us d=0,p=0 1 = 20
Case 2: If a = bd , (or logb a = d) log2 1 = log2 20
log2 1 = 0 log2 2
 If p > -1, then T(n) = Θ(logp+1 n ) log2 1 = 0
T(n) = Θ( logp+1 n )
 If p = -1, then T(n) = Θ(log log n )
= Θ( log0+1 n )
 If p < -1, then T(n) = Θ ( )
= Θ( log n )
= Θ(log n ) Required Answer
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)a: Here a = 2 , b = 2

𝒏
T(n)¿ 𝟐 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 21 log2 2 1
Case 2: If a = bd , (or logb a = d) 2=2 log2 21 1
a = bd 1 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 2 1 = 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)a: Here a = 2 , b = 2

𝒏
T(n)¿ 𝟐 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 21 log2 2 1
Case 2: If a = bd , (or logb a = d) 2=2 log2 21 1
a = bd 1 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 2 1 = 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)a: Here a = 2 , b = 2

𝒏
T(n)¿ 𝟐 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

 If p < -1, then T(n) = Θ ( )

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)a: Here a = 2 , b = 2

𝒏
T(n)¿ 𝟐 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n ) T(n) = Θ( logp+1 n )

 If p = -1, then T(n) = Θ(log log n ) = Θ( log0+1 n )

 If p < -1, then T(n) = Θ ( ) = Θ(log n ) Required Answer

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)b: Here a = 4 , b = 2

𝒏
T(n)¿ 𝟒 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
4 22 log2 4 2
Case 2: If a = bd , (or logb a = d) 4=4 log2 22 2
a = bd 2 log2 2 2
Case 3: If a < bd , (or logb a < d) Case 2 2 = 2
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)b: Here a = 4 , b = 2

𝒏
T(n)¿ 𝟒 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
4 22 log2 4 2
Case 2: If a = bd , (or logb a = d) 4=4 log2 22 2
a = bd 2 log2 2 2
Case 3: If a < bd , (or logb a < d) Case 2 2 =2
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)b: Here a = 4 , b = 2

𝒏
T(n)¿ 𝟒 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

 If p < -1, then T(n) = Θ ( )

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(i)b: Here a = 4 , b = 2

𝒏
T(n)¿ 𝟒 𝑻 ( )+ n2 and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n2 = n2
Give us d=2,p=0 4 = 22
Case 2: If a = bd , (or logb a = d) log2 4 = log2 22
log2 4 = 2 log2 2
 If p > -1, then T(n) = Θ(logp+1 n ) log2 4 = 2
 If p = -1, then T(n) = Θ(log log n ) T(n) = Θ( logp+1 n )

 If p < -1, then T(n) = Θ ( ) = Θ( log0+1 n )

= Θ( log n ) Required Answer

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n1
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 1 , p = -1
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 21 log2 2 1
Case 2: If a = bd , (or logb a = d) 2=2 log2 21 1
a = bd 1 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 2 1 = 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n1
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 1 , p = -1
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 21 log2 2 1
Case 2: If a = bd , (or logb a = d) 2=2 log2 21 1
a = bd 1 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 2 1 = 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n1
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 1 , p = -1
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

 If p < -1, then T(n) = Θ ( )

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n1
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 1 , p = -1
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

T(n) = Θ(log log n )
 If p = -1, then T(n) = Θ(log log n )
= Θ(log log n )
 If p < -1, then T(n) = Θ ( )
= Θ(log log n )
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(iii): Here a = 16 , b = 4

𝒏 and
T(n)¿ 𝟏𝟔 𝑻 ( )+
𝟒
f(n) = O(n) = = n2
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 2 , p = -2
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
16 42 log4 16 2
Case 2: If a = bd , (or logb a = d) 16 = 16 log4 42 2
a = bd 2 log4 4 2
Case 3: If a < bd , (or logb a < d) Case 2 2 = 2
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(iii): Here a = 16 , b = 4

𝒏 and
T(n)¿ 𝟏𝟔 𝑻 ( )+
𝟒
f(n) = O(n) = = n2
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 2 , p = -2
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
16 42 log4 16 2
Case 2: If a = bd , (or logb a = d) 16 = 16 log4 42 2
a = bd 2 log4 4 2
Case 3: If a < bd , (or logb a < d) Case 2 2 = 2
Master Theorem for Dividing Functions : Divide and Conquer

Example 2(iii): Here a = 16 , b = 4

𝒏 and
T(n)¿ 𝟏𝟔 𝑻 ( )+
𝟒
f(n) = O(n) = = n2
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 2 , p = -2
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n )

 If p = -1, then T(n) = Θ(log log n )

 If p < -1, then T(n) = Θ ( )

Master Theorem for Dividing Functions : Divide and Conquer

Example 2(iii): Here a = 16 , b = 4

𝒏 and
T(n)¿ 𝟏𝟔 𝑻 ( )+
𝟒
f(n) = O(n) = = n2
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 2 , p = -2
Case 2: If a = bd , (or logb a = d)

 If p > -1, then T(n) = Θ(logp+1 n ) T(n) = Θ ( )

 If p = -1, then T(n) = Θ(log log n ) =Θ()
 If p < -1, then T(n) = Θ ( )
=Θ()
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
1 21 log2 1 1
Case 2: If a = bd , (or logb a = d) 1<2 log2 20 1
a < bd 0 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 3 0 < 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
1 21 log2 1 1
Case 2: If a = bd , (or logb a = d) 1<2 log2 20 1
a < bd 0 log2 2 1
Case 3: If a < bd , (or logb a < d) Case 3 0 < 1
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0

Case 3: If a < bd , (or logb a < d)

 If p ≥ 0, then T(n) = Θ(nd logp n)

 If p < 0, then T(n) = Θ(nd)

Master Theorem for Dividing Functions : Divide and Conquer

Example 3(i): Here a = 1 , b = 2

𝒏
T(n)¿ 𝑻 ( )+ n and
𝟐
a ≥ 1, b > 1, f(n) = O(n) f(n) = O(n) = n = n1
Give us d=1,p=0

Case 3: If a < bd , (or logb a < d)

 If p ≥ 0, then T(n) = Θ(nd logp n) T(n) = Θ(nd logp n)

 If p < 0, then T(n) = Θ(n ) d = Θ(n1 log0 n)

= Θ( ) Required Answer
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n3
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 3 , p = -1
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 23 log2 2 3
Case 2: If a = bd , (or logb a = d) 2<8 log2 21 3
a < bd 1 log2 2 3
Case 3: If a < bd , (or logb a < d) Case 3 1 < 3
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n3
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 3 , p = -1
Case 1: If a > bd , (or logb a > d)
a bd or logb a d
2 23 log2 2 3
Case 2: If a = bd , (or logb a = d) 2<8 log2 21 3
a < bd 1 log2 2 3
Case 3: If a < bd , (or logb a < d) Case 3 1 < 3
Master Theorem for Dividing Functions : Divide and Conquer

Example 3(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n3
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 3 , p = -1

Case 3: If a < bd , (or logb a < d)

 If p ≥ 0, then T(n) = Θ(nd logp n)

 If p < 0, then T(n) = Θ(nd)

Master Theorem for Dividing Functions : Divide and Conquer

Example 3(ii): Here a = 2 , b = 2

𝒏 and
T(n)¿ 𝟐 𝑻 ( )+
𝟐
f(n) = O(n) = = n3
a ≥ 1, b > 1, f(n) = O(n)
Give us d = 3 , p = -1

Case 3: If a < bd , (or logb a < d)

T(n) = Θ(nd)
 If p ≥ 0, then T(n) = Θ(nd logp n)
T(n) = Θ(n3)
 If p < 0, then T(n) = Θ(nd)