Dynamic Programming

Dynamic Programming

Assembly Line Scheduling

 Dynamic Programming

1. Steps in Dynamic Programming

1. Characterize structure of an optimal solution.
2. Define value of optimal solution recursively.
3. Compute optimal solution values either top-down with
caching or bottom-up in a table.
4. Construct an optimal solution from computed values.
 Dynamic Programming

Elements of Dynamic Programming

 Optimal substructure
 Overlapping subproblems
 Optimal Substructure

 Show that a solution to a problem consists of making a

choice, which leaves one or more subproblems to solve.
 Suppose that you are given this last choice that leads to an
optimal solution.
 Given this choice, determine which subproblems arise and
how to characterize the resulting space of subproblems.
 Show that the solutions to the subproblems used within the
optimal solution must themselves be optimal. Usually use
cut-and-paste.
 Need to ensure that a wide enough range of choices and
subproblems are considered.
 Optimal substructure varies across problem domains:
 1. How many subproblems are used in an optimal solution.
 2. How many choices in determining which subproblem(s) to use.

 Informally, running time depends on (# of subproblems overall)  (# of

choices).
 How many subproblems and choices do the examples considered
contain?
 Dynamic programming uses optimal substructure bottom up.
 First find optimal solutions to subproblems.

 Then choose which to use in optimal solution to the problem.

Does optimal substructure apply to all optimization
problems? No.
Applies to determining the shortest path but NOT the
longest simple path of an unweighted directed graph.
Why?
 Shortest path has independent subproblems.
 Solution to one subproblem does not affect solution to
another subproblem of the same problem.
 Subproblems are not independent in longest simple path.
 Solution to one subproblem affects the solutions to other
subproblems.
 Example:
 Overlapping Subproblem

The space of subproblems must be “small”.

The total number of distinct subproblems is a
polynomial in the input size.
 A recursive algorithm is exponential because it solves the
same problems repeatedly.
 If divide-and-conquer is applicable, then each problem
solved will be brand new.
 Dynamic Programming

Longest Common Subsequence

Document diffing is one of the most prominent uses

of LCS.
 It forms basis of data comparison programs such as
the diff utility, and has applications in computational
linguistics and bioinformatics.
S1 = {B, C, D, A, A, C, D}
S2 = {A, C, D, B, A, C}
Then, common subsequences are {B, C}, {C, D, A, C}, {D,
A, C}, {A, A, C}, {A, C}, {C, D}, ...
Among these subsequences, {C, D, A, C} is the longest
common subsequence.
A brute force approach :-enumerate all subsequence of X
and check each subsequence to see it is also subsequence
of Y
There are 2m subsequence of X
This approach requires exponential time
 example

Let us take two sequences:

1.first sequence =“ACABD”, Second Sequence=“CBDA”
“CD” of length 2
2.LCS for input Sequences
“ABCDGH” and “AEDFHR” “ADH” of length 3.
3.LCS for input Sequences
“AGGTAB” and “GXTXAYB”
“GTAB” of length 4.
Consider the strings "PQRSTPQRS" and
"PRATPBRQRPS". What is the length of the longest
common subsequence?
A 9
B 8
C 7
D 6
Consider two strings A = “qpqrr” and B = “pqprqrp”.
Let x be the length of the longest common
subsequence (not necessarily contiguous) between A
and B and let y be the number of such longest
common subsequences between A and B. Then x +
10y = ___.
(A) 33 (B) 23 (C) 43 (D) 34
The LCS is of length 4. There are 3 LCS of length 4
“qprr”, “pqrr” and qpqr
 A sub-sequence of a given sequence is just the given sequence with some
elements (possibly none or all) left out. We are given two sequences X[m]
and Y[n] of lengths m and n respectively, with indexes of X and Y starting
from 0.
We wish to find the length of the longest common sub-sequence(LCS) of
X[m] and Y[n] as l(m,n), where an incomplete recursive definition for the
function l(i,j) to compute the length of The LCS of X[m] and Y[n] is given
below:
l(i,j) = 0, if either i=0 or j=0
= expr1, if i,j > 0 and X[i-1] = Y[j-1]
= expr2, if i,j > 0 and X[i-1] != Y[j-1]
(A) expr1 ≡ l(i-1, j) + 1
(B) expr1 ≡ l(i, j-1)
(C) expr2 ≡ max(l(i-1, j), l(i, j-1))
(D) expr2 ≡ max(l(i-1,j-1),l(i,j))
 A sub-sequence of a given sequence is just the given sequence with some elements (possibly
none or all) left out. We are given two sequences X[m] and Y[n] of lengths m and n,
respectively with indexes of X and Y starting from 0.
We wish to find the length of the longest common sub-sequence (LCS) of X[m] and Y[n] as
l(m,n), where an incomplete recursive definition for the function I(i,j) to compute the
length of the LCS of X[m] and Y[n] is given below:
l(i,j) = 0, if either i = 0 or j = 0
= expr1, if i,j > 0 and X[i-1] = Y[j-1]
= expr2, if i,j > 0 and X[i-1] ≠ Y[j-1]
The value of l(i, j) could be obtained by dynamic programming based on the correct
recursive definition of l(i, j) of the form given above, using an array L[M, N], where
M = m+1 and N = n + 1, such that L[i, j] = l(i, j).Which one of the following
statements would be TRUE regarding the dynamic programming solution for the
recursive definition of l(i, j)?
A) All elements of L should be initialized to 0 for the values of l(i, j) to be properly computed.
B)The values of l(i,j) may be computed in a row major order or column major order of L[M,
N].
C) The values of l(i, j) cannot be computed in either row major order or column major order of
L[M, N].
D ) L[p, q] needs to be computed before L[r, s] if either p < r or q < s.