Anusnlhan aemed o bo Untwrslily |

racuify of Bngineertngamd TechnoloshTERL SAha "0'

APRIL-2024
MID SEMESTER EXAMINATION,
ALGORITHM DESIGN-2 (CSE4l31)
Programme: B.Tech. (CSE/CSIT/CDS/CIOT/CAIML/CCS)
Time: 2 Hours
Semester: 4h Full Marks: 30 Marks
Ques.
Course Outcome Taxonomy
evel Nos. 2+2+2
1(a), 1(b),
CO1: understand the network flow problem and apply L3,)L4,L5
1(c),
it to realworld problems. 2+2+2+2
L3, L4, L5, 2(a),2(b). +2+2+2+
CO2: -distinguish between computationally 2(c), 3(a).
3(b). 3(c),. 2+2
tractable and intractable problems.
define and relate class-P, class-NP and class 4(a), 4(b),.
NP-complete, PSPACE, PSPACE-complete 4(c).
given problem in NP, define an
appropriate certificate and the verification algorithm.
5(a),5(b), 2+2+2
CO3: understand approximation algorithms and apply
L4,L5,LG
5(c)
this concept to solve problems.
CO4: understand local search techniques and apply
this concept to solve problems.
this
CO5: understand randomization and apply
concept to solve problems.
algorithmic
CO6: identify and apply an appropriate
and explain the
approach to solve a problem
(L2),
challenges tosolve it.
Remembering (L1), Understanding
*Bloom's taxonomy levels: Evaluation (L5), Creation (L6)
Application (L3), Analysis (LA), carries equal mark.
question
Answer all questions. Each
flow-network G(V, E), where V = {s, a,
1. Consider a the edges in E
g, h, i, t and capacities of
b, c, d, e, f, c(s,c)=2, cla,d)=2, cb,d) =1,
are cls,a)=2, cs,b)=3, cle,h)=2,
c(b,e)=1, clc,e)=2, clc,f=1, cld,g)=3,
cg.t)=2, ch,t)=3, cli,t)=2 where s and t are
cf,i)=1, respectively.
the source and sink node
Compute the maximum flow from sto t of the above 2
(a) algorithm.
graph using Ford-Fulkersonand find the capacity of 2
(b) Show the minimum-cut
minimum cuts.
Page 1 of 4
2eof4 2 Give a)a c) b) Givthe a) c)
b)
ete Set-Cover.
the (v7, (v2, andGiven a Give and (Gi) ) C3 they ProvideDefne
problems
(a)
GivenNP-Complete?
given v9)}, v8), set optimization Is Is = are Giv e
Convex-Hul
problem to and
l tol omawsxi:munm 7). of . j A
proper of
a CC x1
clauses
V the a P, (edges 2, . ), 4,5. bipartite
graph? can (v3, graph reduction formal A classified that tor 8). (2, 6.
all CGA A Cz x class sim pl e M,
" (3, Given 6), 7,
v5),a edges G(V, Sel e ch graph
example
em VertexCover
Justity (v4, problern satisfiabe
version
V0 :C= ar e
NP=H¡rd. reduction and matching 6)} a(2, 8, 9;
EE) algotithm C3
X
asnot
justity two
maxi of
the mal
iG=(4
8 ), of
where V satisfiable NP-Hard.
necessar1ly of (3 ,
of verfices
cenario your v5), VOV0,
by ((v1, of
statement ? algorithm2 your
problem.
possible P1 pairs
four
all 5),
(v5, "Set-Cover" examples HoW fivematching(3,6),(3, E)
ingwhere answer. and has
v6), v2), the is
of for
? Ch= selection,"
augmenting pairs the
size (v1,Vertex-Cover in does E
(v6, for Explain whyNP. to
we x, reduce vertices, M,9), sets =
4 v9), set NP-Hard of it (4, (0, {
problem. both
of
one can possible v7), verticesof2 V0 differ vertices' ={0,9), V
=
paths 6), (0, 9),
solve
(v7, (v2, decision V0. Sorting find (4, (1, 1,
of from
its a for v8), v3), Sp as a(1, 7)) 5), 2,
2 2 2 2 2 2

5.

(a) problem
Consider(c) (b)

For Greedy-Load-Balance
(n,m,tl1...n) operators
{ etc.(Hint:
) state". problemempty Cnpy
inding on tiles from Giproblem".
quantified-satisfiable ve features,
any of one
possible for set start tiles (everyone
A[1],A(21,"Alm]
return
job. work of the set assign
set let = j Tiwith set a space. space.
the a
Model lies to 3-satifiable formal i.e.-"solve
the TiA) Mi 1, = sequence
of the
Provide same lowermabove - job be and 0 noconditions, in match The
tile
Ti jobs thisstarting The tch has
j
asboundgiven
machines A) to
+machine a do
n
{A() problem
a tj
machine assigned problem pre of objective
primary the problem? arbitrary
critical the U = moves from fin al ofproblem.
maximum on
approximation H
with
for initial number"Given
must the
Mi all challengea is statement
view
optimal
do
the
machines configuration
as
state, athatgiven to
fromn a How
instances
on
atleast
processing
minimum plagoalnning leadsinitial ofplace 3x3l
this to itfor
assignment
algorithm, Mi
state, to the the board is of
lower the
the mink the state
problsetem. 8-puzzlnumbers usi
e ng and
8)
decision-
different
bound. amount
time Ik goal and with8 given
isOne 2 of tthehe one
of 2
2
a What wil! be the resulting makespan of running this
greedy algorithm on a given sequence of nFm(m-)+]
iobs with processing time t l for 1(=i(=n-1 and
t,=m. for m=4 indentical machines?
() Compare the resulting assignment in Q5(b) with the 2
optimal assignment which can be obtained by
assigning thelargest job to one of the machines and
evenly spreading the remaining jobs over the other m
1machines
*** End of Ouestions ** *