1010043316

(ANALYSIS & DESIGN OF

ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

CHAPTERS 4 – DYNAMIC PROGRAMMING

1. Introduction

 Dynamic programming, like the divide-and-conquer method, solves problems by

combining the solutions to sub problems.

 Divide-and-conquer algorithms partition the problem into independent sub problems,

solve the sub problems recursively, and then combine their solutions to solve the original
problem.

 In contrast, dynamic programming is applicable when the sub problems are not
independent, that is, when sub problems share sub problems.

 In this context, a divide-and-conquer algorithm does more work than necessary,

repeatedly solving the common sub problems.

 A dynamic-programming algorithm solves every sub problem just once and then saves
its answer in a table, thereby avoiding the work of recomputing the answer every time
the sub problem is encountered.

 The development of a dynamic-programming algorithm can be broken into a sequence of

four steps.

1. Characterize the structure of an optimal solution.

2. Recursively define the value of an optimal solution.

3. Compute the value of an optimal solution in a bottom-up fashion.

4. Construct an optimal solution from computed information.

2. Principle of Optimality

 The dynamic programming algorithm obtains the solution using principle of optimality.

 The principle of optimality states that “in an optimal sequence of decisions or choices,
each sub-sequence must also be optimal”.
DEPARTMENT OF COMPUTER ENGINEERING Page | 1
*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

 When it is not possible to apply the principle of optimality it is almost impossible to

obtain the solution using the dynamic programming approach.

 The principle of optimality: “If k is a node on the shortest path from i to j, then the part
of the path from i to k, and the part from k to i, must also be optimal.”

3. Calculating the Binomial coefficient

Consider the problem of calculating binomial coefficient

Suppose 0 ≤ k ≤ n. if we calculate ( ) directly by

function C(n, k)

if k=0 or k=n then return 1

else return C(n-1, k-1)+ C(n-1, k)

Many of the values C(i, j), i < n, j < k, are calculated over and over, for example the
algorithm calculates C(5,3) as sum of C(4,2) and C(4,3) Both these intermediate results
require us to calculate C(3,2). Similarly the value of C (2,2) is used several times.

4. Making change problem using dynamic programming

4.1. Algorithm

function coins (N )

{Gives the minimum number og coins needed to make change for N units. Array
d[1..n] specifies the coinage: in the example there are coins for 1, 4 and 6 units.}

array d[1..n] =[1, 4, 6]

array c[1..n,0..N]

for i←1 to n do c[i,0]←0

DEPARTMENT OF COMPUTER ENGINEERING Page | 2

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

for i←1 to n do

for j←1 to N do

c[i,j]←if i=1 and j < d[i] then +∞

else if i=1 then 1+ c[1,j-d[1]]

else if j < d[i] then c[i-1,j]

else min (c[i-1,j],1+ c[1,j-d[i]]

return c[n,N]

 We need to generate table c[n][N]. Where,

 n= number of denominations. Here we are having 3 denomination so n=3.

 N= number of units that you need to make change. Here we need change of 8 units so
N=8

 To generate table c[i][j] use following steps:

 Step-1: Make c[i][0]=0 for 0 < i ≤ n

 Step-2: Repeat step-2 to step-4 for remaining matrix Values

if i=1 then c[i][j] = 1+c[1][j-d1], here d1=1

 Step-3: if j<di then c[i][j] = c[i-1][j]

 Step-4L: otherwise c[i][j] = min(c[i-1][j],1+c[i][j-di]

Example: Denominations: d1=1, d2=4, d3=6. Make a change of Rs. 8.

c[i][j]=

DEPARTMENT OF COMPUTER ENGINEERING Page | 3

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

We need minimum C[3][8]=2 coins for change

DEPARTMENT OF COMPUTER ENGINEERING Page | 4

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

5. Assembly line scheduling

 Each line has n stations: S1,1, . . . , S1,n and S2,1, . . . , S2,n

 Corresponding stations S1, j and S2, j perform the same function but can take different
amounts of time a1, j and a2, j

 Entry times are: e1 and e2; exit times are: x1 and x2

 After going through a station, can either:

 stay on same line at no cost, or

 transfer to other line: cost after Si,j is ti,j , j = 1, . . . , n – 1

Steps:

f* : the fastest time to get through the entire factory

fi[j] : the fastest time to get from the starting point through station S i,j

f* = min (f1[n] + x 1, f2[n] + x2)

Base case: j = 1, i=1,2 (getting through station 1)

f1[1] = e1 + a1,1

f2[1] = e2 + a2,1

General Case: j = 2, 3, …,n, and i = 1, 2

Fastest way through S1, j is either:

 the way through S 1, j - 1 then directly through S1, j, or

 f1[j - 1] + a1, j

 the way through S 2, j - 1, transfer from line 2 to line 1, then through S 1, j

DEPARTMENT OF COMPUTER ENGINEERING Page | 5

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

f2[j -1] + t2,j-1 + a1,j

6. Knapsack problem

 We need to generate table V(1…n,0…W) where, n= number of objects. Here n=5

DEPARTMENT OF COMPUTER ENGINEERING Page | 6

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

 W= capacity of knapsack. Here W=11

 To generate table V[i][j] use following steps

 Step-1: Make V[i][0] = 0 for 0 < i ≤ n

 Step-2: if j < ωi then take V[i][j] = V[i-1][j]

 Step-3: if j ≥ ωi then take V[i][j] = max(V[i-1][j], V[i-1][j-ωi]+Vi)

 Solution: V[i][j]=

DEPARTMENT OF COMPUTER ENGINEERING Page | 7

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

DEPARTMENT OF COMPUTER ENGINEERING Page | 8

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

DEPARTMENT OF COMPUTER ENGINEERING Page | 9

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

7. Shortest path

7.1. Algorithm

function Floyd(L[1..n, 1..n]) :array [1..n, 1..n]

array D[1..n, 1..n]

D←L

for k ← 1 to n do

for i ← 1 to n do

for j ← 1 to n do

D[i,j] ← min(D[i,j], D[i,k]+ D[k,j])

return D

 We construct a matrix D that gives the length of shortest path between each pair of
nodes.

 The algorithm initializes D to L, that is, to the direct distances between nodes. It then
does n iterations, after iteration k, D gives length of the shortest paths that only use nodes
in {1, 2… k} as intermediate nodes.

DEPARTMENT OF COMPUTER ENGINEERING Page | 10

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

 After n iterations, D therefore gives the length of shortest paths using any of the nodes in
N as an intermediate node.

 If Dk represents the matrix D after k th iteration it can be implemented by

 Dk [i,j] = min(Dk-1 [i,j], Dk-1 [i,k]+ Dk-1 [k,j])

 We use principle of optimality to compute length from i to j passing through k.

7.2. Example

DEPARTMENT OF COMPUTER ENGINEERING Page | 11

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

8. Matrix chain multiplication

 Matrix table M[i][j] stores cost of multiplications

 Matrix chain orders table S[i][j] stores order of multiplication

 To generate M[i][j] use following steps

Step-1 if i = j then M[i][j] = 0

Step-2 if i < j then M[i][j] = min( M[i][k]+M[k+1][j]+P [i-1]* P [k]* P [j]) with

i≤k< j

8.1. Example

Here dimensions are p0=5,p1=4,p2=6,p3=2,p4=7

DEPARTMENT OF COMPUTER ENGINEERING Page | 12

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

DEPARTMENT OF COMPUTER ENGINEERING Page | 13

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

PRINT_OPTIMAL (A, i, j)

1. If i=j

2. print Ai

3. Else

4. print ‘(‘

5. PRINT_OPTIMAL (A, i, k)

6. PRINT_OPTIMAL (A, k+1, j)

7. print ‘)‘

Here we are having 4 matrices let us denote it as A1,A2,A3,A4 so Calling

PRINT_OPTIMAL (A, 1, 4)

Note :For values of k see matrix S[i][j]

DEPARTMENT OF COMPUTER ENGINEERING Page | 14

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

9. Longest chain sub-sequence

9.1. Algorithm

LCS-LENGTH(X, Y)

m ← length[X]

n ← length[Y]

for i ← 1 to m

do c[i, 0] ← 0

for j ← 0 to n

do c[0, j] ← 0

for i ← 1 to m

do for j ← 1 to n

do if xi = yj

then c[i, j] ← c[i - 1, j - 1] + 1

b[i, j] ← "↖ "

else if c[i - 1, j] ≥ c[i, j - 1]

DEPARTMENT OF COMPUTER ENGINEERING Page | 15

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

then c[i, j] ← c[i - 1, j]

b[i, j] ← "↑"

else c[i, j] ← c[i, j - 1]

b[i, j] ← "←"

return c and b

 We need to generate table c(1..m, 1..n) where m=length of string S1 and n= length of
string S2

 b[i][j] stores directions like(←, ↑, ↖ )

 To generate table c[i][j] use following steps

 Step-1: Make c[i][0]=0 and c[0][j]=0

 Step-2: if xi = yj then c[i,j] ← c[i-1,j-1]+1 and b[i,j]←”↖ ”

 Step-3: else if c[i-1,j] ≥ c[I,j-1] then c[i,j] ← c[i-1,j] and b[i,j] ← ”↑”

 Step-4 else c[i,j] ← c[i,j-1] and b[i,j] ← ”←”

9.2. Example

Find any one Longest Common Subsequence of given two strings using Dynamic
Programming.

S1=abbacdcba S2=bcdbbcaa

Solution:

C[i][j]=

DEPARTMENT OF COMPUTER ENGINEERING Page | 16

*Proprietary material of SILVER OAK UNIVERSITY
 1010043316
(ANALYSIS & DESIGN OF
ALGORTIHM)
LECTURE COMPANION SEMESTER: 5 PREPARED BY: PARTH S WADHWA

C[i][j]=

Longest common subsequence = bcdca, Length =5

DEPARTMENT OF COMPUTER ENGINEERING Page | 17

*Proprietary material of SILVER OAK UNIVERSITY