0% found this document useful (0 votes)
24 views32 pagesDAA Unit-3
Daa handwritten notes
Uploaded by
arunmiracle143Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
24 views32 pagesDAA Unit-3
Daa handwritten notes
Uploaded by
arunmiracle143Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 32
Dynamic progrouoting:
Dynamic Programming Wa fechatque thal breaks
the problems Into Subpsoblems , and Saves the sesalt
for future purposes So thad we do not need te Compute
the west again. The Subproblems ave opliwtzed to
Optinfce lho Overall Solution TS Known as optim
Swhstructuae propert4-
Solve
“Me mofn Use of Aynamic programming aS to Solve
OptimFation problem ce
Herve optimfzation prollems mean thd we ave
eeag to find ow the Minimum of the mofinan
Solation of aproblerry-
The dyoamte progsamein g quasantees to %rd the
Oprimal solution of a probern Wf the Solution exists
The definition otf dynante Progvammiag Says thad tt
isa techn?que fox. Solving comple x problem by firck
breaking foto a collection of simpler Subproblems ,
Sohfng each Subprobiem just Once, and Ahern Stoaog
the?e solutfon, to aved ro PTR ue compttalions. oni 5
o') 8 3,5, 8) 13, 9)
one not tandemly
J
a ;
The numbers Pn above Series
Cal, s
Aedsatid. Mothemet oally ,we could write eath oF
the -
ss terms using the belew formula:
Fon ya che 2>\
with the hase values Flod-0o,FCD=) TO
Cateulat the Other Numbers , we follous above
Beladfon ship + Foa example FC) G5 the Sur of Fee £15
ce
tohtch Ps equal bo 4.
Hew can we cod culat, £O2e) 8
oe
4) Fie) Fue) (F'5)
eae oy
Lundamenta) cspeck
9s a
ch Stata thes tH
The prnctple of optimality
of dynamic programning whe
ution *0 a. dysante optinkakien
optimad solutigng to Ft
a s problem Can
be found by He ‘combing the
Sub problems: Of knapsack problem
~? Uwe core Given ‘n' no.of objects usith wetghtS wy
And prottts p; where ‘f/ varies from 4 fo ny and
So a knapsack wth Capackey’M -
> The problem is, we have to fill the bag with
the help of ‘n' objects and the resulting proltt hac
to be maximum:
~? This Problemy 2s Similar te ordinary KnapsSack problen
bat we may not take a fraction of an object
Maxtmtye = PX
1
Subjected to 2 wix; =, rthen CPi) Or i
elfmPnated. (wemove Object ustth less Pag a
More wefoht).
Finding Xt values:
Se ~_
4) set xqco, if (rw) Es”
a) Else Set xed, and (p;) = (P-Po ,W-Wn)
3) nen4 . repeat from Step s:uatil n= o
VY) Solve the solutfon for o/; knorpSac peeve
dynamfe P¥Oqvamming-
CPt, e2,P3) = {rn 2,5)
(t,22,63) = (2,34) M=6 N23
initially eee {ceo}
Sit CPywid= (1/25
fer), (o+2))= Curd
[+1
2S 3st SiS =pmerge Sand $ mh
Sis £C9@), C12) '
S
StS Cpr,wid= (2,3)
SV = (@42, 045), Cita, 243))
St= (ea), (3,52)
itl
tet
S365 so iets S: and sy
p
S'= { Co), Cd 621393,5) }
Sw
1 = (P33) = (S45
Sis (@rs, org), (145 2H), (445,344)
3 = (6a), (eb), (UH. (OD)
(ats, 5+#))
sit! atl 3 2 .
2S 2S = {merge S and Sif
S i Coe, (1,2, (213) Gep-ca) Ces)
[con
With the help purging
Tule toe Con vemove CPj wi) that
is @s) (wv heve profit is less
Weight ts high) .
2) (203) Cs
74) ies Ca, 3) al
Now we have to choose CPi) pofs wh
that is
B
Sia { Ceo) Gr
Pech wsetght
less than (ov) equal to (msc).
CP) = C6,¢)
| for wetght = 6, profit is 6.
Xj Values are © AN Pafrs shortest path: (Ts alse known as, leud- wary
a
we have to find the shortest path eres tal
pats of vertices in a rected graph'al-
> The all pairs Shorstest path problem oi Beer
th of Shortest
& matryfx ‘A’ Such that ACi,j) %s the | nj
Path from i toj-
'
wv) wepsesent the length of
™ Let us take ake;
tej geing “through DOU as
Shostest path trom
index greater than K+ we obtain the formula
KI .
Ga, Aidt akc nia |
t
of
5
AG) = min f A
IgKen
— 2
At first the output malsfx %S Same as given
ref MalsTx Of the qraph: After that oukpur
cost
mahix wt rf
arrix will be updated with all Vert?ces “Kas
the fntermedtat: Vertex.
_s The time complexity of thes algorithm i
hese 'V 7S the numbex of vertices 8 the graph:
$ ov),
—
eC. Bes mee
A All pag -
poss Shore
path lacs? Hen;
Pp prt thn;
Algor,
~~
x Allpaths Ccost 74,0)
SET isn. penn =
y pg ete Wie cork adjacency mabrixn OF a
Jerh uty
A path + n vertfees 5 47,9] the Cost of a shortest
Vertex ‘fF tevetex'j" + Cost(?, 3) = 078 for
y
iN
=O
In
3
for T 224 te x de
tor jz) tendo
Alt 5Je = cost Ct, fF] J eopy cost fntoa
Lor Kr=1 ton do
for fr=) to ndo
tor diz] wn do
Ali SJi = min (acy al, AL KI+ aves) s tritially Care
ee
. iad
A=,13 © 2-0
3/5 WM o4
4lt 2 2 O
Now we find ea* wnat is at Cmeans *ci,g) Shortest
Path fron 4! by through Vertex kK’ Cvia'k se
oe Starts from 4 ten
4 . ee KI s
(G5, AO a CeAy
a a ees ses >)
A 2)3) = min (8,3), PID + (3) )
-™in { 2,40)
AG) =R
a* (2,4) = min( a (2,4) , A (a+ (U4
= min (oo, 8+)
= min (a, 15)
A‘ (24) = \S
A'C3/2) = min (¥G2 ® (3A) + (2)
= min (x, 5 +3)
A'(3,2) 2 Min(0,3)2R fran DOE ee
| A C34) = min A°C34), APC) fA CL)
; = min (1, 5+7)
j = min (1, 10)
Aaa) = 4
4! ;
Cora) = min Ca Cad, C4, D4 OD
= min (2, 243) =min(oo,5)
Aad =F
Aenea ae
Ca) min (a C4 ih Cat ® C3
= min (co, 2 +00)
\ a
A(4#3) =? Vand
So er ae
/ Fesultant mab tse Ale. 2) ¢ o 2 Is]
3/5 30 \
Row te fs, 4 ee ie(i2- Seo 2
ee ah
oe ifo 35 4
A= +{ 8° 215
3/5804
tes [ican eed sO)
AC ud = mia (Aus Az) + A 3)
=™in(o, 2 .
AYCy3) = 5 “~ iene
Ay
Cu4) = min CAcu4 AC 2+ A (a4)
Gry 3415)
= mn,
PUP) =F © IEEE S*C ENE Sa
ae
GD = Min C43, , “&'C3,
2 44'02,0)
pn ey S+8)
= min (5,16)
ae @)= 5
AC 3g) =
aan (alg, a'(3,+ A 4)
min ( 4, e+ Is)
=min (1, 23)
LO)*
A C4 “a
= Min (a'C4,d, Al C429 +A (Ad)
min( 2, s+ ED)
= min iii
Car) =
AA ¢ 4,3) =min CAC D, A! C124 A C278) )
= Mn (0, s+ 2)
= Min Cx, #)
C4) =F
So
7 Cesaltant Mali ?
Now ier A> (2) = mia (2 euad ys PCD)
=min(a, 518)
= min (3,13)
Cp) = a
Aug) = tin Ca*Cugy, A CUDt+ A342)
= min( 4, Sti)
min Gre)
PaO= 6
5
Cae ee (Pam, Dt? C2)
=min(g,2+5)
= ™in( 8,4)
Perma
A> (2,4) = min CCD ) (a D+ ea)
aman (is, 2+!) , .
= min(is,3)
a? Gy) = 3
A> Cait) = mio Ca’ CHD, RCH S)TA'CS/D)
= min Car, 475)
= min (2,12)
Ai) = > :
2 es 5
C4, = min Car Ces , A Card Oth -
2min( 5, T+s)
= min (5,15)
ACE) =F BS
Resultant mah?x is ade
we
pute:
yao!
yor ue
Dd Find At
os
ACen poms:
2/5 023
3{3 6 C4
ee)
A Cad) = min (A8Cu2) , a3(yad +845)
= min ( 3, 645)
ean GUD)
AT Cw) 2S)
A‘ 1/3) = min CA*CDy 2+ PED
= min( 5, 6+)
= min (6,13)
atcd= 5
At (2,1) = min LGD PP (2 Curd)
amin CF, Oat D)
= min(3S)
Ae F
1
At (4,1) = pin (7,0, Peau
mio (5, i+2)
min (5,3)
At(3))~ > —
aC |, 8) 2 min (PO) Bong tA C4ed
= min (27 2+T)* min (210) = *
\
\ ae
A* (3,2 = min (0242), AP (3,44 04.3)
=o Ge, +5)
= min C8,c)
aed G
Resultant matitx ish (3
~ S
pe
NO Ww
bopw
Onacoras
>
w a
wnor
Ona
OoOr=P
EE
ay Teuettng ales Person's problem;
Let @ be rected graph clenoted by (u,E> Where WV!
enotes Ba eer cr meester edges.
—
he edges Ove given Wth they cost ey
"RF S© fer all Pand 7.
—
J =00 PF these So no edge extic
7? the Daueliag ScJesman problem %s to Mod a tour of
e.0
Wietm uns Costs
DA kous of G& i a Brected cycle thal Foclucke every
Vertex Pn 'v'.
7 The cost of the tous {5 the sam of the cost of ty
Rdacs on the tous.
“2 The tour 5 -the shortest path thad starts and end
cd the Same vertex Gayar
Procedumy »—
ee
—>let the functfon BCA, V-{Aq) Ys the total
length of
the tous teenfoattng aS 4) let wR PF) be the Cost of
A
two VerHees Tand 5
—> lee Vi, Un: ua be ‘the Sequence of Vertes followe
Yo optimal tous. Then Ti Naem ny entice emibes a shos test
path —f¥onn Vi to vn Uk
fh passes through each vertex
Exactly once. | >
gperson problem
Overy paiy
visits every hey
youk
as Stance pe
And yetums to the startfng ety:
Starting Vartex
Set of elments ese) olf woaye :
[LP B+ 4 (8, fc,04)] lez e2 ee
@ MO 9 C046, irae =39
wa , .
CB, 5¢
J ce OUD. malt cee
(LCP to ceamp) ie
” CB, + 3 CPA }
Ce fp ao.
) min { wee, * 4Co,g>}
V4 2 ig
I Corfeq
> = (wc YO 4 4 Ce, »|
=e q OS
d+4 26
3 Cer {Bo} |
Mn} Cuccay + 9 CHAD], = MRO] C A+ ay
Letsnd+ 4 (0 f8y)}})- win [ced + q Coe)
let 5= 21
d C482) = wal wen) 18149]
= 1248 = Qe OO Tl
wow,
BCO- Lect) = mm ((rcoimre acerter), = wa} caret = a9,
<)>
(wernt acetsp) = C24 -@ J
9 (8, fe = rn (wor 1)
T 134G 219
J Cer $03) = we [we Coe) + 4g C899)
4At+&=1¢5
Shortest PosKble vole Fs
© DW —-O— -® = a
Shoxgrest Afstane - ‘23
ee an
@) pune
C 13 © es
cet Retranetrey dest
Wgni
P®oblen, Statimenp-
“> To destgn 2 System -thed Fs Composed of Several
doves Connected Fn Sesto.
> Let Ww be the TelPabtttiq ef device Df. Then the
PeBobsettey OF the enite sqsiem %s 7 ¥_
> Even Pe ‘inclvPtural deutes are Vere velfable, the
SelFoubrleey Of th. SySter may not be very Good,
Hence A ©
"TS MPsPrable to dupl
Co Pe
SSS mathe
PoroJlel thro
Fak dite. Cmul ple
Same eviee type Ore connected Pr
egh the USe of Qu tche CPx cuFt s
Sr stage ss contains me copies OF the cleufe Or.’
‘Then the ProbabPle ty thad mo
have a malfunction Fs
me
Ci-a) te
Hence the velfabr itty of sage “S" became s
Tee = | Gap
hngieg
>it [2] > > Sf}
eee
°
Sexes:
‘nt dleukes DF Leh en , Connected %a
Qa: 5)
Staged Slage2 — Ste -$ ua
By
eee! Dy
Dn
°
mudiple dees connected % paral! 9
lack Stage. Pasrgtng ube Cdemfhance vals):
7 . h that
> T+ ¢! contains Cay,c7) and Cre) SYS the:
saat be elfatnated
. e eltntngted.
Ty eaa and CG7=% , then Cg, ¢9) con
x
move cost).
Cremove deutce withe less welfabrl?ty and 2
Uppes bound of each di
CE |etee 3
ee
evtees CD02, PS asTkty
Pe ererges
Cost values Ceicr,c3)= C30,!5,9°) and velabrlers
e cost of the total sys
Design a g-stage system with ¢
C¥1, 32,43) = Cong, 0-8, 0-5) - Th
Should not exceed \os-
C=loy
zc = C+ tls
= 3074 T5+30
=ce = 65
oo
Remairving Amount (G35) — (05-65 -
$o
0
By we pea De as
c Lay
- om Yori
~| '094+30- =
= = = L ¥S.
30
es Ug 9
30
ones) } CR,cde
iC 1,09}
[s"= {me ii “S| and ly |
I . o>
Si = f Cog, 30) f Gen
* Co-99, 1-1 0"
= { Co-99', 60)} an
—6rol
(0-4, 30) Co. WwW cf [oo
o-99
c a Emerge of Sa and a verdes]
> = ae I= Cony
: CoD
1 ©-865, 6) (0-g504772)}
2 dors) >
Ss = 40-8928, 79) f eae
0-96
, oP 3
S = f ¢0-42, 45) (0 T9975 F5) (0-865 60) | eeerss
eee ae) I~ G-0.6F
on 1=(.a93
Sa ieee) (o- 864 60) Co. sqas, 75) f) — 0.008
0-992 wv Scidiaa ies or bei oamiarts na
Optima | Binary Search Trees'- Copst)
IO b
Problem Desoiplion:
eee na 4
Renttiiers Suchthal
> Lee Lar.ar an} be a set of
Atos Let pcey be the proba! tty
gcful Search-
Ber) be the probabritty of tunsucce
iy
Epes) © + Sqew obtain a tree wWeth mPemun
eesti) o jae)
—— cost of Computing Von Can be given as
c
~CTen) = cory CCR 4 Plape tee. + Prt Lot ait +!
Step a;
re Pply the Folloustng fosmula:
* CCP, 3) - mint CIGr, ed HECK, BG + wh, 3)
P @ @ 5
Von
TeKed Coch et) rece sy )4 wet, 3)
MO, 99_ a
SAE WEE F=1) 4 POH 405)
SCID)
e-
4) @raras say) - Cde, 74, Fnt, wh?le )
Slag) - C3.3,11 ‘Ylorgy = (2/3, Vt)
i-Tfeo
Too Ty Tas Tas Toy or
nN
Moo )
CCI T)<20
€ (0) = 6 atD<0
* (0) = gee) = 2 wl D=4q)
(oye) =o 133
z ° 4 Just joes
Joy ‘Ne 23 Taq a Css eae
aa Viz \T23
= aes e Be 1s = 7a
To, (wai SS) VOD leak og wi a)= 70x
W (0,1) - WlO,D4 pls Ge1 Ta | ¥ 0
C34)=3 Fr
weet 343-9 :
Clo) = mtn {Cocos + eda
Oc k cy sia
eed bo GoD=
g TON. kn4 cad
+
Sw Cy29 OOK P24 4¢
5 3+ 34
fame. “$(0,3) = OY Gd =e
G2) = men ( eo anne
OT Regl 04Gb 001,25
keen Noy
= min § ©+0
W (0,4) =/6
¥ Co, ¢)= =
"23 (2,2) = WC, 2) + PCD)4 94.03)
— 1D =
= tti4)
aS,
CC2,3) min {
acke3
C22) 4 (3, a wl2, 3)
K-23
as
a omin fot 8 3 =3
eae
( « ws
SED, wCLLWNDED
*CS8) 5 w0a,3>9 PO 4Q)
BAete)
=3
fey q) « pa _ Cece Bs) HOA
—
Horse 8S
bre
ey
Wt.
i. ot
ak 2 8
Tos
WCO2)~ wo, Ye P2490)
SS+Bere ia
C0, 2) - mtn eles 7
OeKken J ecepse ev >I. ep.
bene ¢ (914 ¢(a,2)
2 ren git ce Tee 7
8
VCo2) stet
ail)
~ wus) = W(t P49)
= F4ler eg
8
CCuS) = mito Coyne 2% ) 4 wth)
eres
cCyd+ c(3,9
Kra,8 >
> 34qel
- {are
7 A
omeerengmenet=
Years
g lovtect pale Fron. ies
FO 3 » wt
ie ins) a we 1, 3)+ PCG) + 4 CG)
— Fr lrlet
“Cua = a ck + <@, ae) }
C12) + Bre) | 4eie)
C3) 4 “Cy JJ
= ore
see ole 0)
(2x0.
= g
Listes 84 =\4
v2
ek on
Bg ear
23,4
Tou
Vaca d= CO D+ PG) 144)
= Ig +h 2
CCe,4) min) ECO D+ CCL 4D
5 On K <4 SCO en Clas
4)
Gara) make, é
Sy wW
1724, 5 CC me ye eG)
= men ) 0 +19
Sr eB _ ms ae
G4 2 mt
ae Zs40 25 = 16416 = 52
(5 te ee (0,9)
=O
are 0'O@ &@
64-5
FN en §9
SO - 4 SC) q)=3
wes) laa T=
=0
¥CG,G)=0 Algoaftnn OBST (p,4.™)
Woo F
ZG Even Atsifny Tdenfiers arcar<--
