I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
,
p
p
.
901
~
908
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
15
.i
1
.
p
p
9
0
1
-
9
0
8
901
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Deep
in
telligence
for sus
taina
ble
farmin
g
:
a
swa
rm
-
e
mpo
wered
da
ta a
na
ly
tics architect
ure
K
ira
n M
un
i
s
wa
m
y
P
a
nd
ura
ng
a
1
,
Ro
o
pa
s
hree
H
ej
j
a
j
i R
a
ng
a
na
t
ha
s
ha
rm
a
2
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
G
o
v
e
r
n
me
n
t
E
n
g
i
n
e
e
r
i
n
g
C
o
l
l
e
g
e
H
a
ssa
n
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
sv
e
sv
a
r
a
y
a
T
e
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
e
l
a
g
a
v
i
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
A
r
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
D
a
t
a
S
c
i
e
n
c
e
,
G
S
S
S
I
n
st
i
t
u
t
e
o
f
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
f
o
r
W
o
me
n
M
y
s
u
r
u
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
s
v
e
s
v
a
r
a
y
a
T
e
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
e
l
a
g
a
v
i
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
24
,
2
0
2
5
R
ev
is
ed
J
an
1
,
2
0
2
6
Acc
ep
ted
J
an
22
,
2
0
2
6
Th
e
in
c
l
u
sio
n
o
f
c
o
m
p
lex
p
a
tt
e
rn
s
o
f
d
a
ta
i
n
p
re
c
isio
n
a
g
ricu
l
tu
re
(P
A)
in
d
u
c
e
s
a
g
re
a
ter
d
e
g
re
e
o
f
c
h
a
ll
e
n
g
e
s
fro
m
th
e
p
e
rsp
e
c
ti
v
e
o
f
c
a
rry
in
g
o
u
t
c
o
n
v
e
n
ti
o
n
a
l
a
n
a
ly
ti
c
a
l
o
p
e
ra
ti
o
n
s.
Alth
o
u
g
h
p
r
o
li
fe
ra
ted
u
se
o
f
a
rti
ficia
l
in
telli
g
e
n
c
e
(AI)
h
a
s
b
e
e
n
n
o
ti
c
e
d
to
y
iel
d
so
m
e
p
ro
m
isi
n
g
re
su
lt
s
t
o
a
d
d
re
ss
su
c
h
issu
e
s,
y
e
t
t
h
e
y
t
o
o
ha
ve
m
a
n
y
sh
o
rtc
o
m
in
g
s.
He
n
c
e
,
t
h
e
c
u
rre
n
t
m
a
n
u
sc
rip
t
i
n
tro
d
u
c
e
s
a
n
i
n
n
o
v
a
ti
v
e
h
y
b
ri
d
AI
sc
h
e
m
e
to
wa
rd
s
e
n
h
a
n
c
in
g
th
e
a
n
a
ly
t
ica
l
o
p
e
ra
ti
o
n
s
n
e
c
e
ss
a
ry
fo
r
d
e
c
isio
n
-
m
a
k
in
g
i
n
sm
a
rt
fa
rm
in
g
.
Th
e
p
ro
p
o
se
d
sc
h
e
m
e
h
y
b
ri
d
ize
s
a
d
e
e
p
n
e
u
ra
l
n
e
tw
o
rk
(DN
N)
wit
h
a
n
o
v
e
l
sw
a
rm
in
telli
g
e
n
c
e
(S
I)
m
o
d
e
l
f
o
r
o
p
ti
m
izi
n
g
t
h
e
p
e
rfo
rm
a
n
c
e
o
f
i
ts
a
d
o
p
te
d
d
e
e
p
lea
rn
in
g
(
DL
)
m
o
d
e
l.
Tes
ted
o
n
a
sta
n
d
a
rd
d
a
tas
e
t
o
f
a
g
r
ic
u
lt
u
re
,
th
e
p
ro
p
o
se
d
m
o
d
e
l
e
x
h
i
b
it
e
d
a
1
0
%
in
c
re
a
se
in
a
c
c
u
ra
c
y
a
n
d
4
0
%
fa
ste
r
re
sp
o
n
se
ti
m
e
wh
e
n
c
o
m
p
a
re
d
with
c
o
n
v
e
n
ti
o
n
a
l
m
a
c
h
i
n
e
lea
rn
in
g
(M
L)
m
o
d
e
ls,
DL
m
o
d
e
ls,
a
n
d
S
I
m
o
d
e
ls.
Th
e
stu
d
y
c
o
n
tri
b
u
tes
to
a
n
o
v
e
l
b
e
n
c
h
m
a
rk
t
o
wa
rd
s ti
m
e
-
e
fficie
n
t
,
sc
a
lab
le,
a
n
d
in
telli
g
e
n
t
a
n
a
ly
ti
c
s o
n
P
A.
K
ey
w
o
r
d
s
:
Ar
tific
ial
in
tellig
en
ce
Dee
p
lear
n
in
g
Ma
ch
in
e
lear
n
in
g
Pre
cisi
o
n
ag
r
icu
ltu
r
e
Swar
m
in
tellig
en
ce
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Kir
an
Mu
n
is
wam
y
Pan
d
u
r
an
g
a
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
G
o
v
er
n
m
en
t E
n
g
in
ee
r
in
g
C
o
lleg
e
Ha
s
s
an
Af
f
iliated
to
Vis
v
esv
ar
ay
a
T
e
ch
n
o
lo
g
ical
Un
iv
er
s
ity
B
elag
av
i
5
9
0
0
1
8
,
Kar
n
atak
a,
I
n
d
ia
E
m
ail:
m
tech
.
k
ir
an
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
Pre
cisi
o
n
ag
r
icu
ltu
r
e
(
PA)
r
e
p
r
esen
ts
a
m
o
d
er
n
ized
ap
p
r
o
a
ch
to
war
d
s
f
ar
m
in
g
m
a
n
ag
em
en
t
wh
er
e
th
e
ag
r
icu
ltu
r
al
in
p
u
ts
ar
e
o
p
tim
ized
u
s
in
g
d
ata
-
d
r
iv
en
t
ec
h
n
o
lo
g
ies
[
1
]
.
T
h
e
id
e
a
o
f
PA
is
to
in
cr
ea
s
e
s
u
s
tain
ab
ilit
y
,
ef
f
icien
cy
,
a
n
d
p
r
o
d
u
ctiv
ity
.
R
ea
l
-
tim
e
d
ata
is
co
llected
f
r
o
m
e
n
v
ir
o
n
m
en
tal
,
wea
th
er
,
s
o
il,
a
n
d
cr
o
p
f
ac
to
r
s
,
a
n
d
u
s
ed
to
m
a
k
e
in
f
o
r
m
ed
d
ec
is
io
n
s
.
Var
io
u
s
k
ey
tec
h
n
o
l
o
g
ies
u
s
ed
i
n
PA
ar
e
au
to
m
ate
d
m
ac
h
in
er
y
an
d
r
o
b
o
tics
,
v
a
r
iab
le
r
ate
tec
h
n
o
lo
g
y
(
VR
T
)
,
d
r
o
n
es
a
n
d
s
atellites,
an
d
lo
ca
tio
n
-
b
ase
d
in
f
o
r
m
atio
n
s
y
s
tem
s
.
T
h
e
r
e
ar
e
v
ar
io
u
s
ap
p
licatio
n
s
f
o
r
PA,
r
an
g
in
g
f
r
o
m
wee
d
co
n
tr
o
l,
liv
esto
ck
m
an
ag
em
en
t,
cr
o
p
y
ield
m
o
n
ito
r
in
g
an
d
f
o
r
ec
asti
n
g
,
p
est
co
n
tr
o
l,
ir
r
ig
atio
n
m
a
n
ag
em
e
n
t,
cr
o
p
m
o
n
ito
r
in
g
,
to
s
o
il
an
d
lan
d
m
ap
p
in
g
.
Mo
s
t
r
ec
en
tly
,
th
er
e
h
av
e
b
ee
n
v
ar
io
u
s
r
ep
o
r
ted
s
tu
d
ies
s
tatin
g
h
ig
h
er
in
v
o
lv
e
m
en
t
o
f
ar
tific
ial
in
tellig
en
ce
(
AI
)
i
n
P
A.
T
h
er
e
ar
e
a
n
u
m
b
er
o
f
s
p
e
cif
ic
p
r
o
b
lem
s
o
r
c
h
allen
g
es
a
s
s
o
ciate
d
with
PA,
wh
ich
ca
n
o
n
ly
b
e
o
p
tim
ally
r
eso
lv
ed
b
y
ad
o
p
tin
g
a
n
AI
-
b
ased
ap
p
r
o
ac
h
,
n
am
el
y
d
ata
o
v
er
lo
ad
a
n
d
co
m
p
lex
ity
,
as
well
as
s
p
atial
an
d
tem
p
o
r
al
v
ar
iab
ilit
y
,
d
e
m
an
d
s
o
f
p
r
e
d
ictiv
e
-
d
ec
is
io
n
-
m
ak
in
g
,
u
n
ce
r
tain
t
y
in
b
io
lo
g
ical
s
y
s
tem
s
,
d
em
a
n
d
s
o
f
r
ea
l
-
tim
e
d
ec
is
io
n
-
m
ak
in
g
,
id
en
tific
atio
n
o
f
d
is
ea
s
es
an
d
p
ests
,
an
d
o
p
tim
izatio
n
o
f
in
p
u
ts
[
2
]
.
He
n
ce
,
AI
h
as
ev
o
lv
e
d
in
PA
f
r
o
m
i)
co
m
p
u
ter
v
is
io
n
to
wa
r
d
s
d
is
ea
s
e
d
iag
n
o
s
is
,
wee
d
d
etec
tio
n
,
an
d
c
r
o
p
m
o
n
ito
r
in
g
;
ii)
clu
s
ter
in
g
an
d
class
if
icatio
n
to
war
d
s
ca
teg
o
r
izati
o
n
o
f
s
o
il
ty
p
e
an
d
f
ield
zo
n
in
g
;
iii)
tim
e
-
s
er
ies
f
o
r
ec
asti
n
g
co
n
tr
ib
u
tin
g
to
wa
r
d
s
y
ield
an
d
wea
th
er
p
r
e
d
ictio
n
.
AI
e
x
is
ts
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
901
-
9
0
8
902
v
ar
io
u
s
f
o
r
m
s
,
an
d
y
et
m
ac
h
in
e
lear
n
in
g
(
ML
)
an
d
d
ee
p
lear
n
in
g
(
DL
)
ar
e
th
e
o
n
ly
AI
v
ar
i
an
ts
th
at
ar
e
f
o
u
n
d
d
o
m
in
an
t
in
th
e
r
esear
ch
d
o
m
ain
.
On
th
e
co
n
tr
ar
y
,
s
war
m
in
tellig
en
ce
(
SI)
is
a
m
u
c
h
less
s
p
o
k
e
n
a
b
o
u
t
ap
p
r
o
ac
h
,
alth
o
u
g
h
it
is
an
i
n
teg
r
al
p
ar
t
o
f
AI
m
o
d
els.
T
h
e
ef
f
ec
tiv
en
ess
o
f
ML
an
d
DL
h
as
b
ee
n
m
u
ch
r
ep
o
r
ted
i
n
m
an
y
s
tu
d
ies;
wh
ile
it
is
in
ter
esti
n
g
to
n
o
te
th
at
SI
m
o
d
els
ar
e
eq
u
ally
p
o
we
r
f
u
l.
Ad
o
p
tio
n
o
f
SI
ca
n
ass
is
t
in
PA
v
ia
c
o
o
r
d
i
n
a
tio
n
with
a
u
to
n
o
m
o
u
s
d
r
o
n
e/r
o
b
o
t
s
war
m
s
.
I
t
ca
n
also
b
e
u
s
ed
f
o
r
m
a
k
in
g
a
d
ec
is
io
n
to
war
d
s
th
e
u
s
ag
e
o
f
wate
r
a
n
d
f
er
tili
ze
r
,
w
h
er
e
th
e
f
o
r
a
g
in
g
b
eh
a
v
io
r
o
f
an
ts
ca
n
b
e
u
s
ed
f
o
r
m
in
im
izin
g
co
s
t.
SI
m
o
d
els
ca
n
also
b
e
u
s
ed
f
o
r
i
d
en
tify
in
g
an
d
r
esp
o
n
d
in
g
to
o
u
tb
r
e
ak
s
o
f
p
ests
wh
er
e
ag
en
t
s
u
s
e
s
war
m
d
y
n
am
ics
to
f
r
am
e
u
p
p
est
b
eh
a
v
io
r
in
o
r
d
er
to
f
o
r
ec
ast
s
p
r
ea
d
p
atter
n
s
o
f
d
is
ea
s
e
o
r
s
o
m
eth
in
g
s
p
ec
if
ic
r
elate
d
to
a
p
lan
t’
s
h
ea
lth
.
Ap
a
r
t
f
r
o
m
th
is
,
it
is
n
o
ted
th
at
h
y
b
r
id
m
o
d
els
to
o
a
r
e
claim
ed
f
o
r
o
p
tim
ized
p
er
f
o
r
m
an
ce
in
PA;
h
o
wev
er
,
s
u
ch
h
y
b
r
i
d
iz
atio
n
s
ar
e
s
ee
n
m
ain
ly
with
ML
an
d
DL
m
o
d
els
an
d
n
o
t
with
SI
m
o
d
els
[
3
]
.
P
r
io
r
to
th
at,
it
is
n
ec
ess
ar
y
t
o
u
n
d
er
s
tan
d
th
e
r
elate
d
w
o
r
k
a
s
s
o
ciate
d
with
PA,
co
n
s
id
er
in
g
v
ar
ied
ca
s
es
o
f
a
g
r
icu
ltu
r
al
p
r
o
b
lem
s
to
r
ea
liz
e
th
e
co
n
tr
i
b
u
tio
n
o
f
e
x
is
tin
g
s
tate
-
of
-
th
e
-
ar
t
AI
m
o
d
els to
war
d
s
im
p
r
o
v
in
g
PA
-
b
ased
o
p
e
r
atio
n
s
.
Dif
f
er
en
t
ty
p
es
o
f
liter
atu
r
e
h
av
e
b
ee
n
r
ev
iewe
d
t
o
u
n
d
er
s
tan
d
th
e
im
p
licatio
n
s
o
f
d
i
f
f
er
en
t
v
ar
ian
ts
o
f
AI
m
o
d
els
to
war
d
s
PA.
A
r
ec
en
t
s
tu
d
y
u
s
in
g
r
an
d
o
m
f
o
r
est
(
R
F)
m
eth
o
d
s
ex
h
ib
its
a
p
o
ten
tial
s
tr
en
g
th
to
war
d
s
s
o
lv
in
g
class
if
icatio
n
p
r
o
b
lem
s
in
PA
q
u
ite
ef
f
icie
n
tly
,
wh
ile
it
ca
n
also
h
an
d
le
r
eg
r
ess
io
n
[
4
]
,
[
5
]
.
Ho
wev
er
,
R
F
m
o
d
els
h
a
v
e
d
e
p
en
d
en
cies
o
f
tr
ee
s
,
wh
ich
co
u
ld
in
c
r
ea
s
e
th
e
co
m
p
u
tatio
n
a
l
ef
f
o
r
t,
wh
ile
th
eir
in
ter
p
r
etatio
n
is
q
u
ite
c
o
m
p
l
ex
in
co
n
t
r
ast
to
a
s
in
g
le
d
e
cisi
o
n
tr
ee
.
Var
io
u
s
r
esear
ch
er
s
h
av
e
also
u
s
ed
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM
)
to
war
d
s
a
d
d
r
ess
in
g
class
if
icatio
n
p
r
o
b
lem
s
in
PA
o
n
v
a
r
ied
f
o
r
m
s
o
f
d
ata
f
o
r
p
est
d
etec
tio
n
,
wee
d
id
en
tific
atio
n
,
an
d
cr
o
p
h
ea
lth
[
6
]
–
[
9
]
.
Alth
o
u
g
h
th
e
y
h
av
e
a
v
er
y
g
o
o
d
g
en
er
aliza
tio
n
p
er
f
o
r
m
an
ce
a
n
d
ar
e
q
u
ite
e
f
f
ec
tiv
e
in
h
ig
h
-
d
im
en
s
io
n
al
s
p
ac
es,
SVM
-
b
ased
m
eth
o
d
s
ar
e
u
n
s
u
itab
le
f
o
r
lar
g
e
-
s
ca
le
d
atasets
.
Ap
ar
t
f
r
o
m
ML
m
o
d
els,
th
er
e
is
an
in
cr
ea
s
in
g
u
s
ag
e
o
f
DL
m
o
d
els
to
o
.
R
ec
en
t
s
tu
d
ies
u
s
in
g
d
ee
p
n
e
u
r
al
n
etwo
r
k
(
DNN)
h
av
e
b
ee
n
f
o
u
n
d
to
a
s
s
is
t
in
p
r
ed
ictiv
e
m
o
d
ellin
g
as
well
as
to
war
d
s
s
o
lv
in
g
n
o
n
-
lin
ea
r
r
elatio
n
s
h
i
p
-
b
ased
c
o
m
p
licatio
n
s
in
PA
[
1
0
]
–
[
1
2
]
.
I
r
r
esp
ec
tiv
e
o
f
its
ex
ten
s
iv
e
ca
p
ab
ilit
ies
to
war
d
s
lear
n
in
g
c
o
m
p
lex
p
a
tter
n
s
,
s
u
ch
m
eth
o
d
h
as
in
cr
ea
s
ed
en
er
g
y
a
n
d
co
m
p
u
tatio
n
al
co
s
t.
An
o
th
e
r
wid
ely
ad
o
p
ted
DL
m
o
d
el
is
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
etwo
r
k
(
C
NN)
,
witn
ess
ed
in
ex
is
t
in
g
s
y
s
tem
wh
ic
h
m
ain
ly
p
r
o
ce
s
s
im
ag
e
-
b
ased
task
s
to
war
d
s
m
ap
p
in
g
c
r
o
p
ty
p
e,
class
if
icatio
n
o
f
wee
d
s
,
an
d
d
etec
tio
n
o
f
d
is
ea
s
es.
T
h
e
m
ajo
r
ity
o
f
C
NN
-
b
ased
ap
p
r
o
ac
h
es
h
av
e
r
ep
o
r
ted
u
n
b
ea
tab
le
class
if
icatio
n
ac
cu
r
ac
y
p
er
f
o
r
m
an
ce
,
an
d
y
et
th
ey
ar
e
ac
tu
ally
n
o
t
s
u
itab
le
f
o
r
n
o
n
-
v
is
u
al
d
ata,
wh
ich
is
o
n
e
o
f
th
eir
d
o
w
n
s
id
es
[
1
3
]
–
[
1
6
]
.
I
t
also
in
v
o
lv
es
co
m
p
u
tatio
n
ally
ex
p
en
s
iv
e
tr
ain
i
n
g
o
p
er
atio
n
s
.
T
h
er
e
ar
e
v
ar
i
o
u
s
s
tu
d
ies
wh
er
e
S
I
h
as
b
ee
n
u
s
ed
to
war
d
s
s
m
ar
t
f
ar
m
in
g
.
So
m
e
o
f
th
e
s
tu
d
ie
s
h
av
e
r
ep
o
r
ted
ly
u
s
ed
an
t
c
o
lo
n
y
o
p
tim
izatio
n
(
AC
O)
to
war
d
s
o
p
tim
al
r
eso
u
r
ce
allo
ca
tio
n
(
p
esti
cid
es/
f
er
tili
ze
r
r
o
u
tes)
as
well
a
s
f
o
r
p
lan
n
in
g
f
ield
p
ath
(
f
o
r
au
to
n
o
m
o
u
s
tr
ac
to
r
s
)
[
1
7
]
–
[
2
0
]
.
Alth
o
u
g
h
AC
O
ap
p
r
o
ac
h
e
s
ar
e
ef
f
icien
t
f
o
r
s
o
lv
in
g
d
is
cr
ete
co
m
b
in
at
o
r
ial
p
r
o
b
lem
s
y
et
th
ey
ar
e
c
h
ar
ac
te
r
ized
b
y
s
lo
w
co
n
v
e
r
g
en
ce
s
p
ee
d
.
An
o
th
er
f
r
eq
u
e
n
tly
u
s
ed
SI
m
o
d
el
is
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
PS
O)
,
wh
ich
is
r
ep
o
r
ted
ly
u
s
ed
f
o
r
h
y
p
er
p
ar
am
eter
tu
n
i
n
g
o
f
M
L
m
o
d
els
as
well
as
o
p
tim
izin
g
y
ield
f
u
n
ctio
n
,
p
la
n
tin
g
s
tr
ateg
ies,
an
d
ir
r
ig
ati
o
n
s
ch
ed
u
les
[
2
1
]
–
[
2
5
]
.
PS
O
-
b
a
s
ed
ap
p
r
o
ac
h
es
ar
e
q
u
ite
s
im
p
lifie
d
to
b
e
im
p
le
m
en
ted
,
an
d
y
et
th
ey
h
a
v
e
s
u
b
-
o
p
tim
al
p
er
f
o
r
m
an
ce
o
n
h
ig
h
-
d
im
en
s
io
n
al
PA
d
ata.
T
h
e
ex
is
tin
g
s
y
s
tem
h
a
s
also
r
ep
o
r
ted
ly
u
s
ed
g
en
eti
c
alg
o
r
ith
m
(
GA)
to
wa
r
d
s
th
e
f
ea
tu
r
e
s
elec
tio
n
p
r
o
ce
s
s
as
well
as
f
o
r
v
ar
io
u
s
p
r
o
ce
s
s
m
an
a
g
em
en
t
in
s
m
ar
t
f
ar
m
in
g
[
2
6
]
–
[
2
8
]
.
GA
-
b
ase
d
ap
p
r
o
ac
h
es
h
a
v
e
ef
f
ec
tiv
e
g
lo
b
al
o
p
tim
izatio
n
p
er
f
o
r
m
an
ce
,
a
n
d
y
et
t
h
ey
ar
e
a
co
m
p
u
tatio
n
ally
ex
p
e
n
s
iv
e
p
r
o
ce
s
s
.
T
h
e
id
en
tifie
d
r
esear
ch
p
r
o
b
l
em
s
ar
e
as
f
o
llo
ws:
i)
th
er
e
ar
e
h
ig
h
er
d
ep
en
d
en
cies
o
f
h
ig
h
-
q
u
ality
d
ataset
f
o
r
AI
m
o
d
els
in
PA
wh
ich
is
q
u
ite
im
p
r
ac
tical
all
th
e
tim
e,
ii)
f
r
eq
u
en
t
ad
o
p
tio
n
o
f
AI
m
o
d
els
in
P
A
also
m
ea
n
s
th
eir
n
atu
r
e
to
b
e
o
f
b
lack
b
o
x
f
o
r
m
wh
ich
r
ed
u
ce
s
r
eliab
ilit
y
an
d
tr
u
s
t
f
o
r
f
ar
m
er
s
to
u
s
e
th
em
,
iii)
th
er
e
is
s
u
b
-
o
p
tim
al
s
ca
li
n
g
o
f
o
p
tim
izatio
n
m
o
d
els
o
f
an
AI
,
esp
ec
ially
wh
en
s
u
b
jecte
d
to
d
iv
er
s
e
an
d
lar
g
e
f
ar
m
in
g
ar
ea
in
r
ea
l
-
tim
e,
an
d
iv
)
it
is
also
ch
allen
g
in
g
task
to
in
teg
r
ate
m
u
ltip
le
AI
tech
n
iq
u
es
in
PA
wh
er
e
v
ar
io
u
s
u
n
c
o
n
tr
o
llab
le
en
v
ir
o
n
m
e
n
tal
f
ac
to
r
s
ex
is
tin
g
in
r
ea
l
s
ce
n
ar
io
.
Hen
ce
,
th
e
p
r
o
p
o
s
ed
s
y
s
tem
ad
d
r
ess
es
th
ese
p
r
o
b
lem
s
m
o
d
u
lar
ap
p
r
o
ac
h
,
wh
ich
in
v
o
lv
es
en
r
ich
in
g
d
ata
q
u
ality
,
s
im
p
lifie
d
an
d
ef
f
ec
tiv
e
f
ea
tu
r
e
s
elec
tio
n
m
eth
o
d
s
,
an
d
a
h
y
b
r
id
i
z
atio
n
o
f
DL
an
d
SI
m
eth
o
d
s
as a
n
o
v
el
AI
tech
n
i
q
u
e.
T
h
e
aim
o
f
th
e
p
r
o
p
o
s
ed
s
tu
d
y
is
to
war
d
s
in
tr
o
d
u
cin
g
a
h
y
b
r
i
d
AI
m
o
d
el
in
teg
r
atin
g
a
DL
m
o
d
el
an
d
an
SI
m
o
d
el
to
war
d
s
en
h
an
c
in
g
th
e
an
aly
tical
o
p
er
atio
n
s
in
v
o
lv
ed
in
g
e
n
er
alize
d
PA
ap
p
licatio
n
s
,
e.
g
.
,
d
ec
is
io
n
-
m
ak
in
g
,
class
if
icatio
n
,
an
d
p
r
o
ce
s
s
in
g
.
T
h
e
co
n
tr
i
b
u
tio
n
s
o
f
th
e
m
o
d
el
a
r
e
as
f
o
llo
ws:
i)
th
e
m
o
d
el
ca
r
r
y
o
u
t
s
elec
tio
n
o
f
p
o
ten
t
ial
r
elev
an
t
attr
ib
u
tes
u
s
in
g
s
im
p
lifie
d
an
d
y
et
e
f
f
icien
t
tr
ee
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
,
ii)
th
e
s
tu
d
y
p
r
esen
ts
an
ab
s
tr
ac
tiv
e
m
o
d
ellin
g
a
p
p
r
o
ac
h
b
y
h
y
b
r
i
d
izin
g
DL
a
n
d
SI
f
o
r
in
n
o
v
ativ
e
an
aly
tical
s
tr
u
ctu
r
e,
iii)
th
e
h
y
b
r
id
m
o
d
el
u
s
es
DNN
f
o
r
p
r
o
ce
s
s
in
g
PA
d
ata
wh
ile
its
h
y
p
er
p
a
r
am
eter
s
ar
e
r
ev
is
ed
u
s
in
g
a
n
o
v
el
m
etah
e
u
r
is
tic
titl
ed
en
h
an
ce
d
lear
n
in
g
an
d
o
p
tim
izatio
n
s
war
m
(
E
L
OS)
,
wh
ich
is
b
ased
o
n
SI
ap
p
r
o
ac
h
,
a
n
d
iv
)
a
n
ex
ten
s
iv
e
s
tu
d
y
ca
r
r
ie
d
o
u
t
to
p
r
o
v
e
e
f
f
ec
tiv
en
ess
o
f
p
r
o
p
o
s
ed
m
o
d
el
b
en
ch
m
ar
k
ed
with
s
tan
d
ar
d
d
ataset
wh
en
co
m
p
ar
e
d
with
f
r
eq
u
en
tly
a
d
o
p
te
d
AI
m
o
d
els.
T
h
e
s
tr
u
ctu
r
e
o
f
t
h
e
m
an
u
s
cr
ip
t
is
as
f
o
llo
ws:
s
ec
tio
n
2
p
r
esen
ts
an
elab
o
r
ated
d
is
cu
s
s
io
n
o
f
th
e
ad
o
p
ted
r
esear
ch
m
eth
o
d
,
wh
ile
r
esu
lt is
d
is
cu
s
s
ed
in
s
ec
tio
n
3
,
an
d
th
e
co
n
clu
s
io
n
is
g
iv
en
i
n
s
ec
tio
n
4.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
in
tellig
en
ce
fo
r
s
u
s
ta
in
a
b
le
fa
r
min
g
:
a
s
w
a
r
m
-
emp
o
w
ered
d
a
ta
… (
K
ir
a
n
Mu
n
is
w
a
my
P
a
n
d
u
r
a
n
g
a
)
903
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
s
tu
d
y
ad
o
p
ts
a
n
o
v
el
m
ath
e
m
atica
l
an
d
a
n
aly
tical
f
r
am
ewo
r
k
with
a
s
o
le
tar
g
et
to
war
d
s
im
p
r
o
v
in
g
t
h
e
an
aly
t
ical
o
p
er
atio
n
ass
o
ciate
d
with
lar
g
e
-
s
ca
le
ag
r
icu
ltu
r
al
d
ata
u
s
in
g
a
h
y
b
r
id
AI
s
y
s
tem
.
T
h
e
s
ch
em
e
in
teg
r
ates
DNN
m
eth
o
d
o
lo
g
y
with
an
in
n
o
v
ativ
e
E
L
OS
f
o
r
d
ev
el
o
p
in
g
a
n
o
v
el.
Fig
u
r
e
1
h
ig
h
lig
h
ts
th
e
ar
ch
itectu
r
al
d
e
s
ig
n
ad
o
p
ted
in
th
e
p
r
o
p
o
s
ed
s
tu
d
y
,
co
n
s
tr
u
cted
with
v
ar
i
o
u
s
co
m
p
o
n
en
ts
with
an
ex
p
licit
s
et
o
f
o
p
e
r
atio
n
s
p
er
f
o
r
m
e
d
b
y
t
h
em
.
T
h
e
m
o
d
el
tak
es
th
e
in
p
u
t
o
f
r
aw
d
ata
f
r
o
m
an
ag
r
icu
ltu
r
al
co
n
tex
t,
wh
ich
is
th
en
s
u
b
ject
ed
to
p
r
ep
r
o
ce
s
s
in
g
,
wh
ile
th
e
s
elec
tio
n
o
f
p
o
ten
tial
f
ea
tu
r
es is
ca
r
r
ied
o
u
t u
s
in
g
a
tr
ee
.
Fu
r
th
er
,
a
h
y
b
r
id
f
o
r
m
o
f
an
AI
m
o
d
el
u
s
in
g
DNN
an
d
n
o
v
el
E
L
OS
is
im
p
lem
en
ted
to
u
n
d
er
tak
e
o
p
tim
al
d
ec
is
io
n
-
m
a
k
in
g
in
PA.
E
n
r
i
c
h
i
n
g
D
a
t
a
Q
u
a
l
i
t
y
T
r
e
e
-
b
a
s
e
d
F
e
a
t
u
r
e
S
e
l
e
c
t
i
o
n
D
e
e
p
N
e
u
r
a
l
N
e
t
w
o
r
k
M
o
d
e
l
S
w
a
r
m
-
b
a
s
e
d
O
p
t
i
m
i
z
a
t
i
o
n
(
E
L
O
S
)
D
a
t
a
R
e
p
r
e
s
e
n
t
a
t
i
o
n
P
r
e
p
r
o
c
e
s
s
i
n
g
E
x
t
r
a
T
r
e
e
s
C
l
a
s
s
i
f
i
e
r
M
L
P
R
e
L
U
P
r
e
d
i
c
t
e
d
c
l
a
s
s
p
r
o
b
a
b
i
l
i
t
y
E
n
t
i
t
y
r
e
p
r
e
s
e
n
t
a
t
i
o
n
F
i
t
n
e
s
s
e
v
a
l
u
a
t
i
o
n
G
r
o
u
p
A
d
a
p
t
a
t
i
o
n
D
y
n
a
m
i
c
s
P
r
e
d
i
c
t
i
o
n
a
n
d
e
v
a
l
u
a
t
i
o
n
Fig
u
r
e
1
.
Ar
c
h
itectu
r
e
o
f
p
r
o
p
o
s
ed
h
y
b
r
id
AI
m
o
d
el
2
.
1
.
E
nrichin
g
da
t
a
qu
a
lity
T
h
is
is
th
e
f
ir
s
t
im
p
lem
e
n
tatio
n
m
o
d
u
le
th
at
en
s
u
r
es
clea
n
er
an
d
co
n
s
is
ten
t
co
llectio
n
o
f
r
aw
in
p
u
t
d
ata
f
r
o
m
d
iv
er
s
e
ag
r
icu
ltu
r
a
l
m
o
n
ito
r
in
g
d
ev
ices.
T
h
e
s
y
s
tem
ca
r
r
ies
o
u
t
a
p
r
e
p
r
o
ce
s
s
in
g
o
p
e
r
atio
n
th
at
co
n
s
is
ts
o
f
h
an
d
lin
g
m
is
s
in
g
v
alu
es,
f
o
llo
we
d
b
y
n
o
r
m
alizin
g
th
e
s
ca
les
o
f
f
ea
tu
r
es.
F
u
r
th
er
e
n
co
d
in
g
is
ca
r
r
ied
o
u
t
f
o
r
ca
teg
o
r
ical
v
ar
iab
les.
C
o
n
s
id
er
an
em
p
ir
ical
r
ep
r
esen
tatio
n
o
f
th
e
r
aw
ag
r
i
cu
ltu
r
al
d
ataset
as
=
{
(
,
)
|
=
1
,
2
,
…
,
}
,
wh
er
e
th
e
v
ar
iab
le
x
i
r
ep
r
esen
ts
[
1
,
2
,
…
,
]
ar
e
all
r
ea
l
-
n
u
m
b
er
-
b
ased
d
ata
d
ep
ictin
g
a
f
ea
t
u
r
e
v
ec
to
r
.
I
t
c
an
also
b
e
c
o
n
s
id
er
ed
as
s
en
s
o
r
y
r
e
ad
in
g
s
(
e.
g
.
,
h
u
m
id
ity
,
te
m
p
er
atu
r
e,
an
d
s
o
il
p
H)
.
T
h
e
s
ec
o
n
d
v
ar
iab
le
is
a
b
in
ar
y
class
lab
el
[
0
,
1
]
d
ep
ictin
g
th
e
p
o
s
s
ib
le
o
u
tco
m
es,
wh
ile
d
an
d
N
r
ep
r
esen
t
th
e
n
u
m
b
e
r
o
f
o
r
i
g
in
al
f
ea
tu
r
es
an
d
th
e
cu
m
u
l
ativ
e
n
u
m
b
er
o
f
in
s
tan
ce
s
.
T
h
e
f
ea
t
u
r
e
is
th
en
p
r
o
ce
s
s
ed
th
r
o
u
g
h
n
o
r
m
aliza
tio
n
as
(
1
)
.
′
=
−
(
1
)
I
n
(
1
)
,
t
h
e
v
ar
iab
le
an
d
r
ep
r
esen
t
th
e
s
tatis
t
ical
m
ea
n
an
d
s
tan
d
ar
d
d
e
v
iatio
n
,
r
esp
ec
ti
v
ely
,
ass
o
ciate
d
with
th
e
ℎ
f
ea
tu
r
e
(
∈
(
1
,
2
,
…
,
)
.
Ad
o
p
tio
n
o
f
n
o
r
m
ali
z
atio
n
co
n
tr
ib
u
tes
to
war
d
s
eq
u
al
em
p
h
asis
an
d
p
ar
ticip
atio
n
o
f
all
v
ar
iab
les
d
u
r
in
g
th
e
m
o
d
el
tr
ain
in
g
.
T
h
is
m
o
d
el
g
e
n
er
ate
s
an
o
u
tco
m
e
o
f
a
h
ig
h
ly
s
tr
u
ctu
r
ed
d
ataset
f
r
o
m
n
o
is
e.
Un
lik
e
co
n
v
en
tio
n
al
p
r
ep
r
o
ce
s
s
in
g
o
p
er
atio
n
s
th
at
co
n
s
id
er
all
f
ea
tu
r
es
in
th
e
s
am
e
m
a
n
n
er
,
th
e
p
r
o
p
o
s
ed
m
o
d
u
le
is
s
p
ec
if
ically
o
p
tim
ized
f
o
r
PA
d
ata,
wh
ich
o
p
er
ates
co
n
s
is
ten
tly
to
war
d
s
th
e
p
r
o
ce
s
s
o
f
f
ea
tu
r
e
s
elec
tio
n
.
2
.
2
.
T
re
e
-
ba
s
ed
f
ea
t
ure
s
elec
t
io
n
T
h
is
m
o
d
u
le
is
r
esp
o
n
s
ib
le
f
o
r
r
etain
in
g
th
e
m
o
s
t
p
o
ten
ti
al
in
f
o
r
m
atio
n
wh
ile
also
ass
is
tin
g
in
m
in
im
izin
g
th
e
d
im
en
s
io
n
alit
y
o
f
PA
d
ata.
T
h
e
p
r
o
p
o
s
ed
s
ch
em
e
u
s
es
an
ex
tr
a
tr
ee
clas
s
if
ier
f
o
r
ap
p
ly
in
g
a
r
an
k
in
g
s
tr
ateg
y
f
o
r
f
ea
tu
r
es
in
o
r
d
er
t
o
im
p
r
o
v
e
m
o
d
el
ac
cu
r
ac
y
an
d
o
p
tim
ize
th
e
co
m
p
u
tatio
n
al
ef
f
icien
cy
.
An
ex
tr
a
tr
ee
class
if
ier
is
a
ty
p
e
o
f
en
s
em
b
led
s
tr
u
ctu
r
e
f
o
r
m
ed
with
an
ar
b
itra
r
y
d
ec
is
io
n
tr
ee
.
T
h
e
s
ig
n
if
ican
ce
v
alu
e
to
war
d
s
all
th
e
in
d
iv
id
u
al
f
ea
tu
r
e
j
is
m
ath
em
atica
lly
ev
alu
ated
as
(
2
)
.
=
∑
∑
△
(
)
.
(
)
−
∈
=
1
(
2
)
I
n
(
2
)
,
it c
an
b
e
n
o
ted
th
at
f
o
r
m
u
latio
n
o
f
s
ig
n
i
f
ican
ce
v
alu
e
is
d
ep
en
d
en
t u
p
o
n
t
h
e
n
u
m
b
e
r
o
f
tr
ee
s
T
,
th
e
s
et
o
f
n
o
d
es
as
s
o
ciate
d
with
tr
ee
t
as
,
th
e
im
p
u
r
it
y
r
ed
u
ctio
n
at
th
e
n
th
n
o
d
e
a
s
(
)
,
an
d
th
e
in
d
icato
r
f
u
n
ctio
n
I
.
T
h
e
p
r
i
m
e
m
o
tiv
e
o
f
th
is
o
p
er
atio
n
is
to
ass
es
s
th
e
im
p
ac
t
o
f
ea
ch
f
ea
tu
r
e
t
o
war
d
s
r
ed
u
cin
g
th
e
u
n
ce
r
tain
ty
d
u
r
in
g
d
ec
is
io
n
-
m
ak
in
g
o
v
er
v
a
r
i
o
u
s
tr
ee
s
.
T
h
e
s
y
s
tem
f
u
r
th
e
r
s
elec
ts
th
e
ℎ
to
p
f
ea
tu
r
es th
at
ar
e
m
ath
e
m
atica
lly
r
ep
r
esen
ted
as
(
3
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
901
-
9
0
8
904
=
(
{
1
,
2
,
…
,
}
)
(
3
)
In
(
3
)
h
i
g
h
lig
h
ts
th
e
s
elec
ted
to
p
f
ea
tu
r
es
b
ased
o
n
th
eir
s
ig
n
if
ican
ce
s
co
r
e.
T
h
e
s
elec
ted
f
ea
tu
r
es
ar
e
r
an
k
ed
,
w
h
ich
is
s
tatis
t
ically
s
ig
n
if
ican
t to
war
d
s
th
e
task
o
f
p
r
ed
ictio
n
.
I
n
co
n
t
r
ast to
th
e
co
n
v
en
tio
n
al
m
eth
o
d
o
f
f
ea
tu
r
e
s
elec
tio
n
u
s
in
g
co
r
r
elatio
n
an
aly
s
is
o
r
u
n
iv
ar
ia
te
f
ilter
in
g
,
th
e
p
r
o
p
o
s
ed
m
o
d
u
le
ca
n
ac
q
u
ir
e
a
s
o
p
h
is
ticated
s
et
o
f
in
f
o
r
m
ati
o
n
p
er
tain
in
g
to
co
m
p
lex
n
o
n
lin
ea
r
in
ter
ac
tio
n
s
am
o
n
g
th
e
in
v
o
lv
ed
v
ar
iab
les.
T
h
e
n
o
v
elty
o
f
t
h
is
m
o
d
u
le
r
e
s
id
es
in
its
ca
p
ab
ilit
y
to
war
d
s
co
m
b
in
in
g
tr
ee
s
with
s
tr
u
ctu
r
al
r
an
d
o
m
n
ess
f
o
r
ef
f
icien
tly
m
an
ag
i
n
g
th
e
h
eter
o
g
en
e
o
u
s
an
d
h
ig
h
-
d
im
en
s
io
n
al
PA
d
ata.
T
h
e
id
ea
is
to
im
p
r
o
v
e
b
o
th
co
m
p
u
tatio
n
al
e
f
f
icien
cy
a
n
d
m
o
d
el
p
er
f
o
r
m
a
n
ce
.
2
.
3
.
Dee
p neura
l net
wo
rk
mo
del
T
h
e
k
ey
g
o
al
o
f
th
is
m
o
d
u
le
is
to
war
d
s
p
er
f
o
r
m
in
g
o
p
tim
ized
lear
n
in
g
o
p
er
atio
n
s
with
ex
ten
s
iv
e
ab
s
tr
ac
tio
n
f
r
o
m
th
e
c
h
o
s
en
f
ea
tu
r
es,
f
o
llo
wed
b
y
p
r
ec
is
e
p
r
ed
ictio
n
.
T
h
is
m
o
d
u
le
co
n
s
is
t
s
o
f
a
d
if
f
e
r
en
t
n
u
m
b
er
o
f
h
id
d
en
lay
er
s
to
w
ar
d
s
tr
an
s
f
o
r
m
in
g
th
e
in
p
u
t
P
A
d
ata
in
to
p
r
o
g
r
ess
iv
ely
s
o
p
h
is
ticated
f
o
r
m
s
o
f
r
ep
r
esen
tatio
n
.
C
o
n
s
id
er
a
r
ed
u
ce
d
f
ea
tu
r
e
m
atr
ix
`
∈
wh
er
e
th
e
v
alu
e
o
f
k
is
m
u
c
h
less
th
an
d
.
T
h
e
s
y
s
tem
th
en
f
o
r
m
u
lates
th
e
m
u
lti
-
lay
er
p
er
ce
p
tr
o
n
w
h
er
e
th
e
in
p
u
t
la
y
er
is
d
ep
icted
a
s
ℎ
(
0
)
=
`
,
h
id
d
en
lay
er
s
ar
e
d
ep
icted
as
f
o
r
=
1
t
o
L
,
ℎ
(
)
=
(
(
)
.
ℎ
(
−
1
)
+
(
)
)
,
an
d
th
e
o
u
tp
u
t
la
y
er
is
d
ep
icted
as
̂
=
(
(
)
.
ℎ
(
)
+
(
)
)
.
I
n
all
th
ese
s
im
p
lifie
d
f
o
r
m
s
o
f
ex
p
r
ess
io
n
,
th
e
v
ar
ia
b
le
s
(
)
an
d
(
)
r
ep
r
esen
t
weig
h
ts
an
d
b
iases
,
r
esp
ec
tiv
ely
,
f
o
r
th
e
lth
lay
er
,
wh
ile
th
e
v
ar
iab
le
(
.
)
r
ep
r
esen
t
s
an
ac
tiv
atio
n
f
u
n
ctio
n
.
T
h
e
s
y
s
tem
also
u
s
es
th
e
s
ig
m
o
id
f
u
n
ctio
n
(
)
=
1
/
(
+
−
)
is
m
e
an
t
f
o
r
m
ap
p
in
g
with
th
e
b
in
ar
y
o
u
tco
m
e
o
f
(
0
,
1
)
an
d
̂
.
I
t
is
co
n
s
id
er
ed
as
f
o
r
ec
asted
p
r
o
b
a
b
ilit
y
o
f
a
class
.
F
u
r
th
er
,
th
e
b
in
ar
y
cr
o
s
s
-
en
tr
o
p
y
lo
s
s
is
m
in
im
ized
in
o
r
d
er
to
tr
ain
th
e
DNN
m
o
d
el,
wh
ich
is
m
ath
em
atica
lly
ex
p
r
ess
ed
as
(
4
)
.
=
−
1
∑
[
.
l
og
(
̂
)
+
(
1
−
)
.
l
og
(
1
−
̂
)
]
=
1
(
4
)
Acc
o
r
d
in
g
to
(
1
)
,
it
ca
n
b
e
n
o
ted
th
at
an
in
tr
icate
r
elatio
n
s
h
ip
am
o
n
g
f
ea
t
u
r
es
is
ca
p
tu
r
ed
b
y
th
e
m
o
d
el,
w
h
er
e
n
o
n
-
lin
ea
r
ity
is
in
tr
o
d
u
ce
d
b
y
th
e
ac
tiv
atio
n
f
u
n
ctio
n
f
r
o
m
ea
ch
lay
er
,
w
h
ile
p
r
o
p
ag
atio
n
o
f
in
f
o
r
m
atio
n
is
ca
r
r
ied
o
u
t
v
ia
weig
h
ted
co
n
n
ec
tio
n
s
th
r
o
u
g
h
th
e
n
etwo
r
k
.
T
h
is
m
o
d
el
g
e
n
er
ates
an
o
u
tco
m
e
o
f
a
tr
ain
ed
m
o
d
el
to
war
d
s
class
if
y
in
g
an
d
f
o
r
ec
asti
n
g
th
e
r
esu
lts
co
n
tex
tu
ally
with
PA
ap
p
licatio
n
s
.
C
o
n
v
en
tio
n
al
ML
m
o
d
els
en
c
o
u
n
ter
ch
allen
g
in
g
s
itu
atio
n
s
to
war
d
s
an
aly
z
in
g
n
o
n
lin
ea
r
p
atter
n
s
in
m
ass
iv
e
an
d
co
m
p
lex
f
o
r
m
s
o
f
PA
d
a
tasets
;
h
o
wev
er
,
th
e
DNN
m
o
d
el
p
r
o
v
id
es
t
h
e
ca
p
ac
ity
f
o
r
DL
an
d
en
h
an
ce
d
ac
cu
r
ac
y
th
at
f
u
r
th
er
co
n
t
r
ib
u
t
es
to
d
y
n
am
ic
ad
ap
tab
ilit
y
.
T
h
e
p
r
o
p
o
s
ed
ar
c
h
itectu
r
e
is
h
ig
h
ly
d
y
n
a
m
ic,
wh
ile
th
e
s
war
m
-
b
ased
in
tellig
en
ce
s
ch
em
e
o
f
E
L
OS
is
u
s
ed
f
o
r
f
in
e
-
tu
n
in
g
its
o
p
er
atio
n
f
o
r
p
er
f
o
r
m
a
n
ce
-
d
r
i
v
en
an
d
d
ata
-
e
f
f
icien
t m
o
d
ellin
g
.
2
.
4
.
Swa
r
m
-
ba
s
ed
o
ptim
i
z
a
t
io
n
(
E
L
O
S)
T
h
e
p
r
im
e
g
o
al
o
f
E
L
OS
i
s
to
war
d
s
o
p
tim
izin
g
th
e
le
ar
n
in
g
co
n
f
ig
u
r
atio
n
a
n
d
ar
ch
itectu
r
al
p
ar
am
eter
s
o
f
th
e
p
r
io
r
DNN
m
o
d
el
f
o
r
r
ed
u
cin
g
th
e
r
esp
o
n
s
e
tim
e
an
d
in
cr
ea
s
in
g
th
e
ac
cu
r
ac
y
.
T
h
e
f
o
r
m
u
latio
n
o
f
th
is
m
o
d
u
le
i
s
ca
r
r
ied
o
u
t
b
ased
o
n
ec
o
l
o
g
ical
s
war
m
s
ex
h
ib
itin
g
th
ei
r
o
p
p
o
r
tu
n
is
tic
an
d
ad
ap
tiv
e
b
e
h
av
io
r
.
T
h
e
f
o
r
m
u
latio
n
o
f
th
e
in
n
o
v
ativ
e
E
L
O
S
m
o
d
u
le
is
d
esig
n
ed
co
n
s
id
er
in
g
th
e
co
g
n
itiv
e
b
eh
av
io
r
o
f
p
r
ed
ato
r
s
war
m
s
with
an
o
p
p
o
r
tu
n
is
tic
d
y
n
am
ic
o
f
th
e
g
r
o
u
p
.
T
h
e
in
d
iv
id
u
al
n
o
d
es f
in
e
-
tu
n
e
th
eir
b
eh
av
io
r
b
ased
o
n
th
e
d
y
n
a
m
ics
o
f
l
o
ca
l
g
r
o
u
p
s
.
Dif
f
er
en
t
f
r
o
m
c
o
n
v
e
n
tio
n
al
ap
p
r
o
ac
h
es
o
f
SI
,
E
L
OS
d
ep
lo
y
s
a
lear
n
in
g
s
tr
ateg
y
o
f
lo
ca
l
s
u
b
g
r
o
u
p
s
c
o
n
s
id
er
in
g
ad
ap
tiv
e
ch
ar
ac
te
r
is
tics
.
T
h
e
p
r
o
p
o
s
ed
s
ch
em
e
co
n
s
id
er
s
th
e
r
ep
r
esen
tatio
n
o
f
en
titi
es
u
s
in
g
d
iv
er
s
e
p
ar
a
m
eter
s
.
T
h
is
m
o
d
u
le
in
tr
o
d
u
ce
s
a
lear
n
in
g
s
tr
ateg
y
wh
er
e
th
e
ex
p
lo
r
atio
n
o
f
th
e
s
o
lu
tio
n
s
p
ac
e
in
d
if
f
e
r
en
t
r
e
g
io
n
s
is
ca
r
r
ied
o
u
t
b
y
a
s
o
f
twar
e
ag
en
t.
T
h
e
s
y
s
tem
co
n
s
id
er
s
a
s
o
f
twar
e
ag
en
t
∈
d
ep
icts
th
e
s
o
lu
tio
n
v
ec
to
r
th
at
is
r
esp
o
n
s
ib
le
f
o
r
en
co
d
in
g
t
h
e
lear
n
in
g
r
ate,
ac
tiv
atio
n
f
u
n
ctio
n
,
n
eu
r
o
n
s
p
er
lay
e
r
,
an
d
n
u
m
b
er
o
f
h
id
d
en
la
y
er
s
.
An
em
p
ir
ical
s
tatem
en
t
o
f
f
itn
ess
ev
alu
atio
n
(
)
is
r
ep
r
esen
ted
as
(
5
)
.
(
)
=
.
(
)
−
.
(
)
(
5
)
I
n
(
5
)
,
th
e
v
a
r
iab
les
α
an
d
β
r
ep
r
esen
t
th
e
s
ca
lin
g
co
ef
f
icien
t.
T
h
e
s
y
s
tem
f
u
r
th
er
p
r
esen
ts
a
d
y
n
am
ics
o
f
g
r
o
u
p
ad
ap
tatio
n
wh
er
e
t
h
e
p
o
s
itio
n
is
u
p
d
at
ed
b
y
an
a
g
en
t
o
n
th
e
b
asis
o
f
an
o
p
p
o
r
tu
n
is
tic
f
ac
to
r
γ
an
d
∗
r
ep
r
esen
tin
g
a
lo
ca
l
b
est,
wh
er
e
a
s
u
b
g
r
o
u
p
G
co
n
s
is
ts
o
f
all
th
e
ag
e
n
ts
a
.
T
h
e
ex
p
r
ess
io
n
o
f
g
r
o
u
p
ad
ap
tatio
n
o
f
an
ag
e
n
t i
s
as
(
6
)
.
(
+
1
)
=
(
)
+
(
6
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
in
tellig
en
ce
fo
r
s
u
s
ta
in
a
b
le
fa
r
min
g
:
a
s
w
a
r
m
-
emp
o
w
ered
d
a
ta
… (
K
ir
a
n
Mu
n
is
w
a
my
P
a
n
d
u
r
a
n
g
a
)
905
I
n
(
6
)
,
th
e
v
ar
iab
le
δ
r
ep
r
esen
ts
.
(
∗
−
(
)
)
+
an
d
th
e
attr
ib
u
te
[
∈
(
0
,
1
)
]
is
u
s
ed
f
o
r
co
n
tr
o
llin
g
r
ate
o
f
c
o
n
v
e
r
g
en
ce
.
T
h
e
p
ar
a
m
eter
∈
is
n
ea
r
ly
s
im
ilar
to
th
e
Gau
s
s
ian
n
o
is
e
ter
m
(
0
,
2
)
th
at
is
ad
o
p
ted
f
o
r
m
ai
n
tain
in
g
th
e
d
iv
er
s
ity
.
All
th
e
s
o
f
twar
e
a
g
en
t
th
at
ar
e
f
o
u
n
d
to
s
u
b
-
o
p
tim
ally
p
er
f
o
r
m
is
s
u
b
jecte
d
to
re
-
in
itializatio
n
to
a
n
o
v
el
ar
b
itra
r
y
p
o
s
itio
n
af
ter
ev
er
y
R
iter
atio
n
,
wh
er
e
R
r
ep
r
esen
ts
th
e
r
eset
r
u
le.
T
h
is
m
o
d
u
le
g
en
er
ates
an
o
u
tco
m
e
o
f
an
o
p
tim
ized
DNN
m
o
d
el
wh
er
e
v
ar
io
u
s
h
y
p
e
r
p
ar
a
m
eter
s
ar
e
au
to
n
o
m
o
u
s
ly
s
elec
ted
.
Dif
f
er
en
t
f
r
o
m
ex
is
tin
g
SI
ap
p
r
o
ac
h
es,
t
h
is
m
o
d
u
l
e
co
m
b
in
es
s
tr
ateg
ic
r
eset
with
lo
ca
l
lear
n
in
g
f
o
r
r
esis
tin
g
an
y
f
o
r
m
o
f
p
r
em
atu
r
e
co
n
v
er
g
e
n
ce
.
T
h
e
n
o
v
elty
o
f
th
is
m
o
d
u
le
is
th
at
it
y
ield
s
r
o
b
u
s
tn
ess
with
i
n
th
e
PA d
ataset
with
en
h
an
ce
d
ac
cu
r
ac
y
o
f
p
r
e
d
ictio
n
an
d
f
as
ter
co
n
v
e
r
g
en
ce
.
3.
RE
SU
L
T
AND
DI
SCUS
SI
O
N
T
h
e
lo
g
ic
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
is
wr
it
ten
in
Py
th
o
n
,
co
n
s
id
er
in
g
a
n
o
r
m
al
W
in
d
o
ws
6
4
-
b
it
en
v
ir
o
n
m
en
t.
Usi
n
g
J
u
p
y
ter
n
o
teb
o
o
k
,
th
e
d
esig
n
p
r
o
ce
s
s
is
ac
co
m
p
lis
h
ed
u
s
in
g
v
ar
io
u
s
lib
r
ar
ies
an
d
p
ac
k
ag
es,
v
iz.
Nu
m
Py
,
Pan
d
a
s
,
Ma
tp
lo
tlib
,
Seab
o
r
n
,
T
en
s
o
r
Flo
w/Ker
as,
an
d
Scik
it
-
lear
n
.
T
h
e
en
v
ir
o
n
m
en
t
h
as
b
ee
n
s
im
u
lated
o
n
a
n
I
n
tel
C
o
r
e
i7
p
r
o
ce
s
s
o
r
with
1
6
GB
o
f
DDR4
R
AM
.
T
h
e
p
r
o
p
o
s
ed
s
y
s
tem
ex
ec
u
tio
n
is
ca
r
r
ied
o
u
t
o
n
th
e
C
PU
an
d
d
o
esn
’
t
r
eq
u
ir
e
an
y
ex
p
licit
GPU
with
in
th
e
s
e
s
s
i
o
n
tim
e.
Usi
n
g
th
e
s
tan
d
ar
d
d
ataset
[
2
9
]
an
d
a
s
im
ilar
test
en
v
ir
o
n
m
en
t,
th
e
p
r
o
p
o
s
ed
m
o
d
el
(
Pro
p
)
th
at
co
n
s
is
ts
o
f
E
L
OS
o
p
tim
izer
,
DNN,
a
n
d
tr
ee
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
is
co
m
p
ar
ed
with
i)
two
p
o
te
n
tial
ML
m
o
d
els,
v
iz.
R
F
,
an
d
SVM,
ii)
two
p
o
ten
tial
DL
m
o
d
els,
v
iz.
,
b
aselin
ed
DNN
an
d
C
NN,
an
d
iii)
two
p
o
te
n
tial
SI
m
o
d
els,
i.e
.
,
PSO
-
tu
n
ed
DNN
an
d
GA
t
u
n
ed
DNN.
T
h
e
co
m
p
lete
ass
es
s
m
en
t
an
d
b
e
n
ch
m
ar
k
in
g
ar
e
c
ar
r
ied
o
u
t
co
n
s
id
er
in
g
ac
c
u
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F1
-
s
co
r
e
,
an
d
r
esp
o
n
s
e
tim
e.
3
.
1
.
Acc
o
m
pli
s
hed o
utc
o
m
e
T
h
e
o
u
tco
m
e
is
s
h
o
wn
in
T
a
b
le
1
,
wh
ich
ex
h
ib
its
th
e
Pro
p
to
ac
co
m
p
lis
h
m
ax
im
u
m
a
cc
u
r
ac
y
o
f
class
if
icatio
n
(
ac
c
=9
3
%)
wh
ile
it
r
ec
o
r
d
s
m
in
im
ized
d
u
r
atio
n
o
f
r
esp
o
n
s
e
tim
e
(
r
time
=0
.
7
8
s
)
in
co
n
tr
ast
to
all
ex
is
tin
g
m
o
d
els.
T
h
e
p
r
im
ar
y
r
ea
s
o
n
f
o
r
ef
f
icien
cy
a
n
d
m
o
d
el
p
r
ec
is
io
n
is
d
u
e
to
th
e
co
m
b
in
atio
n
o
f
E
L
OS
SI
an
d
ad
o
p
ted
f
ea
tu
r
e
s
elec
tio
n
u
s
in
g
a
tr
ee
.
I
t
ca
n
also
b
e
n
o
ted
th
at
th
e
p
r
o
p
o
s
ed
s
y
s
tem
r
etain
s
a
h
ig
h
er
F1
-
s
co
r
e
(
v
al
=0
.
9
1
)
to
ex
h
ib
it
its
b
ala
n
ce
d
s
p
ec
i
f
icity
an
d
s
en
s
itiv
ity
o
v
er
a
g
r
icu
ltu
r
al
class
es.
Ov
er
all,
th
e
o
u
tc
o
m
e
o
f
th
e
p
r
o
p
o
s
ed
s
tu
d
y
r
ep
r
esen
ts
its
r
ea
l
-
tim
e
ca
p
ab
ilit
y
,
ac
cu
r
ac
y
,
an
d
s
ca
lab
ilit
y
.
T
ab
le
1
.
Nu
m
e
r
ical
o
u
tco
m
es o
f
th
e
s
tu
d
y
M
o
d
e
l
A
c
c
u
r
a
c
y
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
R
e
s
p
o
n
se
t
i
me
(
sec
)
RF
0
.
8
4
0
.
8
2
0
.
8
0
0
.
8
1
0
.
8
9
S
V
M
0
.
8
1
0
.
7
8
0
.
7
6
0
.
7
7
1
.
1
4
B
a
se
l
i
n
e
D
N
N
0
.
8
7
0
.
8
5
0
.
8
3
0
.
8
4
1
.
2
2
C
N
N
0
.
8
8
0
.
8
6
0
.
8
4
0
.
8
5
1
.
3
8
P
S
O
+
D
N
N
0
.
8
9
0
.
8
7
0
.
8
6
0
.
8
6
0
.
9
7
G
A
+
D
N
N
0
.
8
8
0
.
8
6
0
.
8
4
0
.
8
5
1
.
0
1
P
r
o
p
(
E
LO
S
+
D
N
N
)
0
.
9
3
0
.
9
2
0
.
9
1
0
.
9
1
0
.
7
8
3
.
2
.
Dis
cus
s
io
n
T
h
e
ac
co
m
p
lis
h
ed
o
u
tco
m
e
o
f
th
e
s
tu
d
y
ex
h
ib
ited
in
Fig
u
r
e
2
s
h
o
ws
th
at
p
r
o
p
o
s
ed
s
y
s
tem
o
f
f
er
s
ap
p
r
o
x
im
ately
1
0
.
2
%
b
etter
a
cc
u
r
ac
y
in
c
o
n
tr
ast
to
c
o
n
v
e
n
tio
n
al
ML
m
o
d
el
(
R
F
an
d
SVM)
,
ap
p
r
o
x
im
ately
6
.
8
%
in
cr
ea
s
ed
ac
cu
r
ac
y
in
c
o
n
tr
ast
to
tr
ad
itio
n
al
DL
m
o
d
els
(
b
aselin
e
DNN
an
d
C
NN)
,
an
d
ap
p
r
o
x
im
ately
8
.
4
%
o
f
m
ax
im
ized
class
if
icatio
n
ac
cu
r
ac
y
in
co
n
tr
ast
to
co
n
v
en
tio
n
al
SI
m
o
d
el
s
(
PS
O+
DNN
an
d
GA+
DNN)
.
T
h
is
ca
n
b
e
r
ea
lized
f
r
o
m
Fig
u
r
e
2
(
a)
.
Similar
ly
p
r
o
p
o
s
ed
s
ch
em
e
co
n
tr
ib
u
t
es
to
1
2
.
5
%
,
7
.
6
%
,
an
d
9
.
1
% b
etter
p
r
ec
is
io
n
in
c
o
n
tr
ast to
co
n
v
en
tio
n
al
ML
,
DL
,
an
d
SI
m
o
d
els,
as sh
o
wn
in
Fig
u
r
e
2
(
b
)
,
wh
ich
also
s
h
o
ws
r
ec
all
p
er
f
o
r
m
an
c
e
o
f
n
ea
r
ly
s
im
ilar
tr
en
d
s
.
T
h
e
F1
-
s
co
r
e
o
f
th
e
p
r
o
p
o
s
ed
s
y
s
tem
is
wi
tn
ess
ed
to
o
f
f
er
1
1
.
9
%
,
7
.
2
%
,
an
d
8
.
9
%
en
h
an
ce
d
p
e
r
f
o
r
m
an
ce
in
co
n
tr
ast
to
tr
ad
itio
n
al
ML
,
DL
,
an
d
SI
m
o
d
els,
r
esp
ec
tiv
ely
(
Fig
u
r
e
2
(
c)
)
.
Similar
ly
,
Fig
u
r
e
2
(
d
)
s
h
o
ws
th
e
p
r
o
p
o
s
ed
s
ch
em
e
to
ex
h
ib
it
3
4
.
5
%
,
4
2
.
3
%
,
a
n
d
2
7
.
1
% f
aster
r
esp
o
n
s
e
tim
e
in
co
n
tr
ast to
ML
,
DL
,
a
n
d
SI
m
o
d
els.
T
h
e
p
r
im
e
r
ea
s
o
n
f
o
r
th
e
u
n
d
e
r
p
er
f
o
r
m
an
ce
o
f
c
o
n
v
e
n
tio
n
al
m
o
d
els
ca
n
b
e
attr
ib
u
te
d
to
t
h
e
f
ac
t
th
at
ML
m
o
d
els
d
o
n
o
t
o
f
f
e
r
a
b
s
tr
ac
tio
n
lay
er
s
,
wh
ile
th
ey
h
av
e
is
s
u
es
p
er
tain
in
g
to
b
ias/
o
v
er
f
itti
n
g
o
n
v
o
lu
m
in
o
u
s
d
ata.
I
n
t
h
e
c
o
n
v
en
tio
n
al
DL
m
o
d
el
,
th
e
b
as
elin
e
DL
m
o
d
els
d
o
n
'
t
in
clu
d
e
o
p
tim
izatio
n
th
at
r
esu
lts
in
p
latea
u
in
g
o
r
r
estri
cted
co
n
v
er
g
e
n
ce
,
m
ai
n
ly
d
u
e
to
n
o
n
-
o
p
tim
al
h
y
p
er
p
ar
am
eter
s
.
On
th
e
o
th
er
h
an
d
,
th
e
co
n
v
en
tio
n
al
SI
m
o
d
els
ar
e
f
o
u
n
d
to
g
et
en
g
a
g
ed
to
lo
ca
l
o
p
tim
a
th
at
d
o
n
'
t
p
o
s
s
es
s
ad
ap
tiv
e
ch
ar
ac
ter
is
tics
to
war
d
s
b
alan
cin
g
ex
p
lo
itatio
n
an
d
ex
p
lo
r
atio
n
.
On
th
e
co
n
tr
ar
y
,
th
e
p
r
o
p
o
s
ed
s
ch
em
e
in
teg
r
ates
a
DL
m
o
d
el
o
p
tim
ized
b
y
a
n
o
v
el
SI
m
o
d
el
o
f
E
L
OS
,
an
d
tr
ee
-
b
ased
f
ea
tu
r
e
s
elec
tio
n
is
p
er
f
o
r
m
e
d
to
war
d
s
s
tab
ilized
clas
s
if
icati
o
n
an
d
en
h
a
n
ce
s
g
en
er
aliza
tio
n
.
Ap
ar
t
f
r
o
m
th
is
,
it
is
n
o
ted
th
at
th
er
e
is
an
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
901
-
9
0
8
906
ex
h
au
s
tiv
e
u
s
ag
e
o
f
r
u
le
-
b
as
ed
s
tr
u
ctu
r
e
in
co
n
v
en
tio
n
al
ML
m
o
d
els
th
at
r
esu
lts
in
a
lo
n
g
er
d
u
r
atio
n
o
f
ev
alu
atio
n
.
Fu
r
th
er
,
t
h
e
co
m
p
u
tatio
n
in
th
e
DL
m
o
d
el
is
ass
o
ciate
d
with
co
m
p
lex
lay
e
r
s
lack
in
g
o
p
tim
izatio
n
,
r
esu
ltin
g
in
in
cr
ea
s
ed
r
esp
o
n
s
e
tim
e.
I
n
clu
s
io
n
o
f
iter
atio
n
s
with
in
SI
m
o
d
els
is
an
o
th
er
r
ea
s
o
n
f
o
r
its
m
ax
im
ized
r
esp
o
n
s
e
tim
e.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
2
.
B
en
ch
m
ar
k
ed
o
u
tco
m
es
o
f
th
e
s
tu
d
y
of
(
a)
ac
cu
r
a
cy
,
(
b
)
p
r
ec
is
io
n
a
n
d
r
ec
all
,
(
c
)
F1
-
s
co
r
e,
an
d
(
d
)
r
esp
o
n
s
e
ti
me
4.
CO
NCLU
SI
O
N
T
h
e
p
r
o
p
o
s
ed
s
tu
d
y
p
r
esen
ts
a
n
o
v
el
an
aly
tical
a
n
d
co
m
p
u
tatio
n
al
f
r
am
ewo
r
k
t
h
at
is
ca
p
ab
le
o
f
o
p
tim
izin
g
th
e
d
ec
is
io
n
-
m
a
k
in
g
in
PA
b
y
in
teg
r
atin
g
a
n
o
v
el
SI
m
o
d
el
with
a
D
L
m
o
d
el.
T
h
e
k
e
y
co
n
tr
ib
u
tio
n
is
to
war
d
im
p
r
o
v
in
g
th
e
an
aly
tical
o
p
er
atio
n
wh
en
ex
p
o
s
ed
to
a
v
o
lu
m
i
n
o
u
s
PA
d
ataset.
I
t
i
s
n
o
ted
th
at
p
r
e
p
r
o
ce
s
s
in
g
o
v
er
h
ea
d
ca
n
b
e
p
o
ten
tially
co
n
tr
o
lled
b
y
m
in
im
izin
g
ir
r
elev
a
n
t
f
ea
tu
r
es,
wh
ich
is
tak
en
ca
r
e
o
f
b
y
a
d
o
p
tin
g
tr
ee
-
b
ased
s
elec
tio
n
o
f
attr
ib
u
tes.
Fu
r
th
er
,
E
L
OS
is
u
s
ed
f
o
r
f
in
e
-
tu
n
in
g
th
e
DNN
m
o
d
el
th
at
r
esu
lts
in
an
o
p
tim
ized
tr
ain
in
g
m
o
d
el.
T
h
e
lear
n
in
g
o
u
tco
m
e
c
o
n
tr
ib
u
ted
b
y
th
e
s
tu
d
y
s
h
o
ws
th
a
t
co
m
p
u
tatio
n
al
ef
f
icien
cy
is
co
n
tr
ib
u
ted
b
y
th
e
p
r
o
p
o
s
ed
tr
ee
-
b
ased
attr
ib
u
te
s
elec
tio
n
,
wh
ile
h
ig
h
-
q
u
ality
f
lo
w
o
f
in
f
o
r
m
atio
n
is
co
n
tr
ib
u
ted
b
y
t
h
e
ab
s
tr
ac
tiv
e
d
ata
m
o
d
el,
a
n
d
o
p
tim
iza
tio
n
o
f
in
tellig
en
t
h
y
p
er
p
ar
am
eter
s
is
ca
r
r
ied
o
u
t
b
y
E
L
OS.
Ov
er
all,
th
e
p
r
o
p
o
s
ed
s
tu
d
y
s
h
o
wca
s
es
th
at
th
e
d
ep
lo
y
m
en
t
o
f
AI
s
y
s
tem
s
to
war
d
r
ea
l
-
tim
e
d
ec
i
s
io
n
-
m
ak
in
g
is
q
u
ite
f
ea
s
ib
le
i
n
PA
f
o
r
u
n
d
er
ta
k
in
g
r
ea
l
-
tim
e
d
ec
is
io
n
-
m
a
k
in
g
f
r
o
m
th
e
p
er
s
p
ec
tiv
e
o
f
s
m
ar
t
ir
r
ig
atio
n
c
o
n
tr
o
l,
y
ield
p
r
e
d
ictio
n
,
an
d
m
o
n
ito
r
in
g
o
f
cr
o
p
h
ea
lth
.
On
th
e
g
r
o
u
n
d
s
o
f
ac
c
o
m
p
lis
h
ed
r
esu
lts
,
it
ca
n
b
e
s
tated
th
at
th
e
p
r
o
p
o
s
ed
s
ch
em
e
m
ai
n
tain
s
a
g
o
o
d
b
alan
ce
b
etwe
en
in
cr
ea
s
ed
class
if
icatio
n
ac
cu
r
ac
y
an
d
f
aster
r
esp
o
n
s
e
tim
e,
wh
ich
m
a
k
es
it
m
o
r
e
a
p
p
li
ca
b
le
in
f
u
tu
r
is
tic
r
ea
l
-
tim
e
ap
p
licatio
n
s
a
n
d
s
er
v
ices
in
PA.
T
h
e
f
u
tu
r
e
wo
r
k
will
in
v
o
lv
e
f
u
r
th
er
o
p
tim
izin
g
th
e
e
n
tire
p
r
o
ce
s
s
b
y
ad
o
p
tin
g
m
o
r
e
ad
v
an
ce
d
v
er
s
io
n
s
o
f
DL
m
o
d
els
with
eq
u
al
em
p
h
asis
o
n
lo
w
c
o
m
p
u
tatio
n
al
co
s
t
an
d
r
eso
u
r
ce
in
clu
s
io
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
Au
th
o
r
s
s
tate
n
o
f
u
n
d
in
g
in
v
o
lv
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
Dee
p
in
tellig
en
ce
fo
r
s
u
s
ta
in
a
b
le
fa
r
min
g
:
a
s
w
a
r
m
-
emp
o
w
ered
d
a
ta
… (
K
ir
a
n
Mu
n
is
w
a
my
P
a
n
d
u
r
a
n
g
a
)
907
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Kir
an
Mu
n
is
wam
y
Pan
d
u
r
an
g
a
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
o
o
p
ash
r
ee
Hejjaji
R
an
g
an
ath
ash
ar
m
a
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
Au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
t
er
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
i
n
g
s
o
f
th
is
s
tu
d
y
ar
e
a
v
ailab
le
f
r
o
m
th
e
c
o
r
r
esp
o
n
d
in
g
au
th
o
r
,
[
KM
P
]
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
S
a
n
y
a
o
l
u
a
n
d
A
.
S
a
d
o
w
sk
i
,
“
Th
e
r
o
l
e
o
f
p
r
e
c
i
si
o
n
a
g
r
i
c
u
l
t
u
r
e
t
e
c
h
n
o
l
o
g
i
e
s
i
n
e
n
h
a
n
c
i
n
g
su
st
a
i
n
a
b
l
e
a
g
r
i
c
u
l
t
u
r
e
,
”
S
u
s
t
a
i
n
a
b
i
l
i
t
y
,
v
o
l
.
1
6
,
n
o
.
1
5
,
A
u
g
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
s
u
1
6
1
5
6
6
6
8
.
[
2
]
G
.
G
u
p
t
a
a
n
d
S
.
K
.
P
a
l
,
“
A
p
p
l
i
c
a
t
i
o
n
s
o
f
A
I
i
n
p
r
e
c
i
s
i
o
n
a
g
r
i
c
u
l
t
u
r
e
,
”
D
i
s
c
o
v
e
r
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
3
,
n
o
.
1
,
A
p
r
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
4
2
7
9
-
025
-
0
0
2
2
0
-
9.
[
3
]
O
.
A
p
p
i
a
h
e
t
a
l
.
,
“
P
l
a
n
t
e
S
a
i
n
e
:
a
n
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
t
e
mp
o
w
e
r
e
d
mo
b
i
l
e
a
p
p
l
i
c
a
t
i
o
n
f
o
r
p
e
st
s
a
n
d
d
i
se
a
se
man
a
g
e
me
n
t
f
o
r
mai
z
e
,
t
o
ma
t
o
,
a
n
d
o
n
i
o
n
f
a
r
mers
i
n
B
u
r
k
i
n
a
F
a
s
o
,
”
A
g
ri
c
u
l
t
u
re
,
v
o
l
.
1
4
,
n
o
.
8
,
Ju
l
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
4
0
8
1
2
5
2
.
[
4
]
M
.
S
.
D
h
i
l
l
o
n
e
t
a
l
.
,
“
I
n
t
e
g
r
a
t
i
n
g
r
a
n
d
o
m fo
r
e
st
a
n
d
c
r
o
p
m
o
d
e
l
i
n
g
i
m
p
r
o
v
e
s t
h
e
c
r
o
p
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
o
f
w
i
n
t
e
r
w
h
e
a
t
a
n
d
o
i
l
se
e
d
r
a
p
e
,
”
Fro
n
t
i
e
rs
i
n
Re
m
o
t
e
S
e
n
s
i
n
g
,
v
o
l
.
3
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
8
9
/
f
r
sen
.
2
0
2
2
.
1
0
1
0
9
7
8
.
[
5
]
A
.
P
a
n
g
,
M
.
W
.
L.
C
h
a
n
g
,
a
n
d
Y
.
C
h
e
n
,
“
E
v
a
l
u
a
t
i
o
n
o
f
r
a
n
d
o
m
f
o
r
e
s
t
s
(
R
F
)
f
o
r
r
e
g
i
o
n
a
l
a
n
d
l
o
c
a
l
-
s
c
a
l
e
w
h
e
a
t
y
i
e
l
d
p
r
e
d
i
c
t
i
o
n
i
n
S
o
u
t
h
e
a
s
t
A
u
st
r
a
l
i
a
,
”
S
e
n
so
rs
,
v
o
l
.
2
2
,
n
o
.
3
,
Ja
n
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
0
3
0
7
1
7
.
[
6
]
A
.
S
u
mar
u
d
i
n
,
E
.
I
sman
t
o
h
a
d
i
,
A
.
P
u
sp
a
n
i
n
g
r
u
m,
S
.
M
a
u
l
a
n
a
,
a
n
d
M
.
N
a
d
i
,
“
I
mp
l
e
me
n
t
a
t
i
o
n
i
r
r
i
g
a
t
i
o
n
s
y
s
t
e
m
u
si
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
m
a
c
h
i
n
e
f
o
r
p
r
e
c
i
si
o
n
a
g
r
i
c
u
l
t
u
r
e
b
a
se
d
o
n
I
o
T,
”
I
O
P
C
o
n
f
e
r
e
n
c
e
S
e
r
i
e
s:
M
a
t
e
ri
a
l
s
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
0
9
8
,
n
o
.
3
,
M
a
r
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
5
7
-
8
9
9
X
/
1
0
9
8
/
3
/
0
3
2
0
9
8
.
[
7
]
S.
-
H
.
G
o
a
n
d
J.
-
H
.
P
a
r
k
,
“
I
mp
r
o
v
i
n
g
f
i
e
l
d
c
r
o
p
c
l
a
ssi
f
i
c
a
t
i
o
n
a
c
c
u
r
a
c
y
u
si
n
g
G
LC
M
a
n
d
S
V
M
w
i
t
h
U
A
V
-
a
c
q
u
i
r
e
d
i
m
a
g
e
s,
”
K
o
r
e
a
n
J
o
u
r
n
a
l
o
f
Re
m
o
t
e
S
e
n
si
n
g
,
v
o
l
.
4
0
,
n
o
.
1
,
p
p
.
9
3
–
1
0
1
,
2
0
2
4
,
d
o
i
:
1
0
.
7
7
8
0
/
k
j
r
s.
2
0
2
4
.
4
0
.
1
.
9
.
[
8
]
V
.
S
a
x
e
n
a
e
t
a
l
.
,
“
U
t
i
l
i
z
i
n
g
s
u
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
s
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
c
r
o
p
d
i
s
e
a
ses
i
n
p
r
e
c
i
s
i
o
n
a
g
r
i
c
u
l
t
u
r
e
a
d
a
t
a
mi
n
i
n
g
p
e
r
s
p
e
c
t
i
v
e
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
i
n
En
g
i
n
e
e
r
i
n
g
,
v
o
l
.
1
2
,
n
o
.
1
6
s
,
p
p
.
2
8
1
–
2
8
8
,
2
0
2
4
.
[
9
]
R
.
H
.
P
r
i
n
c
e
e
t
a
l
.
,
“
C
S
X
A
I
:
a
l
i
g
h
t
w
e
i
g
h
t
2
D
C
N
N
-
S
V
M
m
o
d
e
l
f
o
r
d
e
t
e
c
t
i
o
n
a
n
d
c
l
a
ssi
f
i
c
a
t
i
o
n
o
f
v
a
r
i
o
u
s
c
r
o
p
d
i
s
e
a
ses
w
i
t
h
e
x
p
l
a
i
n
a
b
l
e
A
I
v
i
s
u
a
l
i
z
a
t
i
o
n
,
”
Fro
n
t
i
e
rs i
n
P
l
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
5
,
J
u
l
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s.
2
0
2
4
.
1
4
1
2
9
8
8
.
[
1
0
]
S
.
Jeo
n
g
,
S
.
J
e
o
n
g
,
a
n
d
J.
B
o
n
g
,
“
D
e
t
e
c
t
i
o
n
o
f
t
o
ma
t
o
l
e
a
f
mi
n
e
r
u
si
n
g
d
e
e
p
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
2
,
n
o
.
2
4
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
2
2
4
9
9
5
9
.
[
1
1
]
J.
K
i
m
,
I
.
D
o
H
a
,
S
.
K
w
o
n
,
I
.
Jan
g
,
a
n
d
M
.
H
.
N
a
,
“
A
sm
a
r
t
f
a
r
m
D
N
N
su
r
v
i
v
a
l
mo
d
e
l
c
o
n
s
i
d
e
r
i
n
g
t
o
m
a
t
o
f
a
r
m
e
f
f
e
c
t
,
”
Ag
ri
c
u
l
t
u
r
e
,
v
o
l
.
1
3
,
n
o
.
9
,
S
e
p
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
3
0
9
1
7
8
2
.
[
1
2
]
C
.
W
e
i
,
Y
.
S
h
a
n
,
a
n
d
M
.
Z
h
e
n
,
“
D
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
f
o
r
p
r
e
c
i
si
o
n
f
i
e
l
d
c
r
o
p
p
r
o
t
e
c
t
i
o
n
,
”
F
ro
n
t
i
e
rs
i
n
Pl
a
n
t
S
c
i
e
n
c
e
,
v
o
l
.
1
6
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
3
3
8
9
/
f
p
l
s
.
2
0
2
5
.
1
5
7
6
7
5
6
.
[
1
3
]
I
.
A
l
-
S
h
o
u
r
b
a
j
i
a
n
d
S
.
D
u
r
a
i
b
i
,
“
I
W
Q
P
4
N
e
t
:
a
n
e
f
f
i
c
i
e
n
t
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
i
r
r
i
g
a
t
i
o
n
w
a
t
e
r
q
u
a
l
i
t
y
p
r
e
d
i
c
t
i
o
n
,
”
Wa
t
e
r
,
v
o
l
.
1
5
,
n
o
.
9
,
A
p
r
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
w
1
5
0
9
1
6
5
7
.
[
1
4
]
Y
.
W
a
n
g
,
Y
.
C
h
e
n
,
a
n
d
D
.
W
a
n
g
,
“
C
o
n
v
o
l
u
t
i
o
n
n
e
t
w
o
r
k
e
n
l
i
g
h
t
e
n
e
d
t
r
a
n
sf
o
r
m
e
r
f
o
r
r
e
g
i
o
n
a
l
c
r
o
p
d
i
se
a
s
e
c
l
a
ss
i
f
i
c
a
t
i
o
n
,
”
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
1
,
n
o
.
1
9
,
O
c
t
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
1
1
9
3
1
7
4
.
[
1
5
]
A
.
O
k
a
y
a
ma,
A
.
Y
a
mam
o
t
o
,
M
.
K
i
m
u
r
a
,
a
n
d
Y
.
M
a
t
s
u
n
o
,
“
U
t
i
l
i
z
i
n
g
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
(
C
N
N
)
f
o
r
o
r
c
h
a
r
d
i
r
r
i
g
a
t
i
o
n
d
e
c
i
si
o
n
-
ma
k
i
n
g
,
”
E
n
v
i
ro
n
m
e
n
t
a
l
M
o
n
i
t
o
r
i
n
g
a
n
d
Ass
e
ssm
e
n
t
,
v
o
l
.
1
9
7
,
n
o
.
2
,
Ja
n
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
6
6
1
-
024
-
1
3
6
0
2
-
1.
[
1
6
]
M
.
A
l
t
u
n
a
n
d
M
.
T
u
r
k
e
r
,
“
I
n
t
e
g
r
a
t
i
o
n
o
f
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
w
i
t
h
p
a
r
c
e
l
-
b
a
se
d
i
ma
g
e
a
n
a
l
y
s
i
s
f
o
r
c
r
o
p
t
y
p
e
ma
p
p
i
n
g
f
r
o
m t
i
me
-
ser
i
e
s
i
m
a
g
e
s,”
E
a
rt
h
S
c
i
e
n
c
e
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
8
,
n
o
.
3
,
S
e
p
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
2
1
4
5
-
025
-
0
1
8
1
9
-
8.
[
1
7
]
A
.
A
sg
h
a
r
i
,
M
.
Ze
i
n
a
l
a
b
e
d
i
n
ma
l
e
k
mi
a
n
,
H
.
A
z
g
o
mi
,
M
.
A
l
i
mo
r
a
d
i
,
a
n
d
S
.
G
h
a
z
i
a
n
t
a
f
r
i
sh
i
,
“
F
a
r
m
e
r
a
n
t
s
o
p
t
i
mi
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
:
a
n
o
v
e
l
m
e
t
a
h
e
u
r
i
st
i
c
f
o
r
s
o
l
v
i
n
g
d
i
scr
e
t
e
o
p
t
i
m
i
z
a
t
i
o
n
p
r
o
b
l
e
ms,”
I
n
f
o
rm
a
t
i
o
n
,
v
o
l
.
1
6
,
n
o
.
3
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
i
n
f
o
1
6
0
3
0
2
0
7
.
[
1
8
]
Z.
Zh
a
n
g
,
P
.
Li
,
S
.
C
h
a
i
,
Y
.
C
u
i
,
a
n
d
Y
.
Ti
a
n
,
“
D
G
A
-
A
C
O
:
e
n
h
a
n
c
e
d
d
y
n
a
mi
c
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
—
a
n
t
c
o
l
o
n
y
o
p
t
i
m
i
z
a
t
i
o
n
p
a
t
h
p
l
a
n
n
i
n
g
f
o
r
a
g
r
i
b
o
t
s,
”
A
g
ri
c
u
l
t
u
r
e
,
v
o
l
.
1
5
,
n
o
.
1
2
,
Ju
n
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
5
1
2
1
3
2
1
.
[
1
9
]
D
.
S
h
a
n
,
S
.
Z
h
a
n
g
,
X
.
W
a
n
g
,
a
n
d
P
.
Z
h
a
n
g
,
“
P
a
t
h
-
p
l
a
n
n
i
n
g
st
r
a
t
e
g
y
:
a
d
a
p
t
i
v
e
a
n
t
c
o
l
o
n
y
o
p
t
i
mi
z
a
t
i
o
n
c
o
mb
i
n
e
d
w
i
t
h
a
n
e
n
h
a
n
c
e
d
d
y
n
a
m
i
c
w
i
n
d
o
w
a
p
p
r
o
a
c
h
,
”
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
3
,
n
o
.
5
,
F
e
b
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
3
0
5
0
8
2
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2252
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
15
,
No
.
1
,
Feb
r
u
ar
y
20
26
:
901
-
9
0
8
908
[
2
0
]
W
.
K
u
a
n
g
,
H
.
W
.
H
o
,
a
n
d
Y
.
Z
h
o
u
,
“
C
P
P
-
D
I
P
:
mu
l
t
i
-
o
b
j
e
c
t
i
v
e
c
o
v
e
r
a
g
e
p
a
t
h
p
l
a
n
n
i
n
g
f
o
r
M
A
V
s
i
n
d
i
sp
e
r
s
e
d
a
n
d
i
r
r
e
g
u
l
a
r
p
l
a
n
t
a
t
i
o
n
s
,
”
Ro
b
o
t
i
c
s
a
n
d
A
u
t
o
n
o
m
o
u
s
S
y
s
t
e
m
s
,
v
o
l
.
1
9
5
,
Ja
n
.
2
0
2
6
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
o
b
o
t
.
2
0
2
5
.
1
0
5
1
9
3
.
[
2
1
]
B
.
Z
h
e
n
g
e
t
a
l
.
,
“
A
n
A
N
N
-
PSO
-
b
a
s
e
d
m
e
t
h
o
d
f
o
r
o
p
t
i
m
i
z
i
n
g
a
g
r
i
c
u
l
t
u
r
a
l
t
r
a
c
t
o
r
s
i
n
f
i
e
l
d
o
p
e
r
a
t
i
o
n
f
o
r
e
m
i
ssi
o
n
r
e
d
u
c
t
i
o
n
,
”
Ag
ri
c
u
l
t
u
r
e
,
v
o
l
.
1
2
,
n
o
.
9
,
A
u
g
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
2
0
9
1
3
3
2
.
[
2
2
]
L.
X
u
,
J
.
Y
o
u
,
a
n
d
H
.
Y
u
a
n
,
“
R
e
a
l
-
t
i
me
p
a
r
a
me
t
r
i
c
p
a
t
h
p
l
a
n
n
i
n
g
a
l
g
o
r
i
t
h
m fo
r
a
g
r
i
c
u
l
t
u
r
a
l
ma
c
h
i
n
e
r
y
k
i
n
e
ma
t
i
c
s mo
d
e
l
b
a
se
d
o
n
p
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
,
”
A
g
r
i
c
u
l
t
u
r
e
,
v
o
l
.
1
3
,
n
o
.
1
0
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
3
1
0
1
9
6
0
.
[
2
3
]
H
.
H
u
a
n
g
,
X
.
C
u
a
n
,
Z.
C
h
e
n
,
L
.
Z
h
a
n
g
,
a
n
d
H
.
C
h
e
n
,
“
A
m
u
l
t
i
r
e
g
i
o
n
a
l
a
g
r
i
c
u
l
t
u
r
a
l
ma
c
h
i
n
e
r
y
sc
h
e
d
u
l
i
n
g
me
t
h
o
d
b
a
se
d
o
n
h
y
b
r
i
d
p
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m,”
A
g
r
i
c
u
l
t
u
r
e
,
v
o
l
.
1
3
,
n
o
.
5
,
M
a
y
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
a
g
r
i
c
u
l
t
u
r
e
1
3
0
5
1
0
4
2
.
[
2
4
]
Z.
X
u
a
n
d
Y
.
S
u
n
,
“
P
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
mi
z
a
t
i
o
n
f
o
r
a
g
r
i
c
u
l
t
u
r
a
l
p
r
o
b
l
e
ms,
”
H
i
g
h
l
i
g
h
t
s
i
n
Bu
s
i
n
e
ss,
E
c
o
n
o
m
i
c
s
a
n
d
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
5
1
,
p
p
.
2
4
5
–
2
5
0
,
F
e
b
.
2
0
2
5
,
d
o
i
:
1
0
.
5
4
0
9
7
/
3
g
4
b
n
p
6
0
.
[
2
5
]
K
.
A
b
b
a
l
,
M
.
E
l
-
A
mr
a
n
i
,
O
.
A
o
u
n
,
a
n
d
Y
.
B
e
n
a
d
a
d
a
,
“
A
d
a
p
t
i
v
e
p
a
r
t
i
c
l
e
sw
a
r
m
o
p
t
i
m
i
z
a
t
i
o
n
w
i
t
h
l
a
n
d
s
c
a
p
e
l
e
a
r
n
i
n
g
f
o
r
g
l
o
b
a
l
o
p
t
i
m
i
z
a
t
i
o
n
a
n
d
f
e
a
t
u
r
e
se
l
e
c
t
i
o
n
,
”
Mo
d
e
l
l
i
n
g
,
v
o
l
.
6
,
n
o
.
1
,
Ja
n
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
m
o
d
e
l
l
i
n
g
6
0
1
0
0
0
9
.
[
2
6
]
W
.
Z
h
a
n
g
e
t
a
l
.
,
“
P
r
e
c
i
s
i
o
n
r
e
g
u
l
a
t
i
o
n
a
n
d
f
o
r
e
c
a
st
i
n
g
o
f
g
r
e
e
n
h
o
u
se
t
o
mat
o
g
r
o
w
t
h
c
o
n
d
i
t
i
o
n
s
u
s
i
n
g
a
n
i
mp
r
o
v
e
d
G
A
-
B
P
mo
d
e
l
,
”
S
u
st
a
i
n
a
b
i
l
i
t
y
,
v
o
l
.
1
6
,
n
o
.
1
0
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
su
1
6
1
0
4
1
6
1
.
[
2
7
]
S
.
S
h
a
r
ma,
B
.
K
.
P
a
t
h
a
k
,
a
n
d
R
.
K
u
mar,
“
A
d
o
p
t
i
n
g
a
n
i
m
p
r
o
v
e
d
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
f
o
r
m
u
l
t
i
-
o
b
j
e
c
t
i
v
e
serv
i
c
e
c
o
m
p
o
s
i
t
i
o
n
o
p
t
i
m
i
z
a
t
i
o
n
i
n
smar
t
a
g
r
i
c
u
l
t
u
r
e
,
”
A
u
st
ri
a
n
J
o
u
r
n
a
l
o
f
S
t
a
t
i
s
t
i
c
s
,
v
o
l
.
5
3
,
n
o
.
5
,
p
p
.
1
1
–
2
5
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
7
7
1
3
/
a
j
s.
v
5
3
i
5
.
1
8
7
4
.
[
2
8
]
J.
Z
h
o
n
g
,
J
.
L
i
n
,
a
n
d
Y
.
Li
n
,
“
O
p
t
i
mi
z
a
t
i
o
n
o
f
c
r
o
p
p
l
a
n
t
i
n
g
s
t
r
a
t
e
g
i
e
s
u
si
n
g
l
i
n
e
a
r
p
r
o
g
r
a
m
mi
n
g
a
n
d
g
e
n
e
t
i
c
a
l
g
o
r
i
t
h
m
u
n
d
e
r
c
o
m
p
l
e
x
l
a
n
d
c
o
n
d
i
t
i
o
n
s,
”
H
i
g
h
l
i
g
h
t
s
i
n
S
c
i
e
n
c
e
,
En
g
i
n
e
e
r
i
n
g
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
4
2
,
p
p
.
4
9
–
5
4
,
M
a
y
2
0
2
5
,
d
o
i
:
1
0
.
5
4
0
9
7
/
q
k
w
j
h
q
8
7
.
[
2
9
]
S
.
K
i
r
u
t
h
i
k
a
a
n
d
D
.
K
a
r
t
h
i
k
a
,
“
I
o
T
-
b
a
s
e
d
p
r
o
f
e
ss
i
o
n
a
l
c
r
o
p
r
e
c
o
m
me
n
d
a
t
i
o
n
sy
s
t
e
m
u
si
n
g
a
w
e
i
g
h
t
-
b
a
se
d
l
o
n
g
-
t
e
r
m
m
e
m
o
r
y
a
p
p
r
o
a
c
h
,
”
M
e
a
s
u
rem
e
n
t
:
S
e
n
so
rs
,
v
o
l
.
2
7
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
a
se
n
.
2
0
2
3
.
1
0
0
7
2
2
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
K
ira
n
Mu
n
iswa
m
y
Pa
n
d
u
r
a
n
g
a
h
a
s
c
o
m
p
lete
d
B
E
(I
n
fo
rm
a
ti
o
n
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
)
in
th
e
y
e
a
r
2
0
0
5
fro
m
S
ri
S
i
d
d
h
a
rth
a
In
st
it
u
te o
f
Tec
h
n
o
lo
g
y
Tu
m
k
u
r,
Affil
iate
d
to
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
iv
e
rsity
,
Be
lag
a
v
i
,
Ka
rn
a
t
a
k
a
,
In
d
ia
a
n
d
M
.
Tec
h
.
(Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
)
in
t
h
e
y
e
a
r
2
0
0
7
fr
o
m
AMC
En
g
i
n
e
e
rin
g
C
o
ll
e
g
e
Ba
n
g
a
lo
re
,
Affili
a
ted
to
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
Un
i
v
e
rsity
,
Be
lag
a
v
i,
Ka
rn
a
tak
a
,
In
d
ia.
He
h
a
s
a
ro
u
n
d
1
4
y
e
a
rs
o
f
tea
c
h
in
g
e
x
p
e
rien
c
e
.
He
is
p
re
se
n
tl
y
w
o
rk
i
n
g
a
s
a
ss
istan
t
p
ro
fe
ss
o
r
in
th
e
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
a
t
G
o
v
e
rn
m
e
n
t
En
g
in
e
e
rin
g
Co
ll
e
g
e
Ha
ss
a
n
-
5
7
3
2
0
1
,
Ka
rn
a
tak
a
,
I
n
d
ia.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
m
tec
h
.
k
iran
@g
m
a
il
.
c
o
m
.
Ro
o
p
a
shre
e
H
e
jja
ji
Ra
n
g
a
n
a
th
a
sha
r
m
a
h
a
s
c
o
m
p
lete
d
BE
(El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
)
a
n
d
M
.
Tec
h
.
(C
o
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
)
fr
o
m
VTU
,
Be
lag
a
v
i,
Ka
rn
a
tak
a
,
In
d
ia
a
n
d
P
h
.
D
.
fro
m
CHRIST
(De
e
m
e
d
to
b
e
Un
iv
e
rsit
y
)
Be
n
g
a
lu
r
u
,
Ka
rn
a
tak
a
,
In
d
ia.
S
h
e
h
a
s
a
ro
u
n
d
1
3
y
e
a
rs
o
f
In
d
u
strial
e
x
p
e
rie
n
c
e
a
n
d
6
y
e
a
rs
o
f
tea
c
h
i
n
g
e
x
p
e
rien
c
e
.
S
h
e
is
c
u
rre
n
tl
y
wo
r
k
in
g
a
s
p
r
o
fe
ss
o
r
a
n
d
h
e
a
d
i
n
t
h
e
De
p
a
rtme
n
t
o
f
Artifi
c
ial
In
telli
g
e
n
c
e
a
n
d
Da
ta
S
c
ien
c
e
a
t
G
S
S
S
In
stit
u
te
o
f
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
y
fo
r
W
o
m
e
n
,
M
y
su
r
u
-
5
7
0
0
1
6
,
Ka
rn
a
tak
a
,
In
d
i
a
a
n
d
su
p
e
rv
isi
n
g
6
P
h
.
D
.
re
se
a
rc
h
sc
h
o
lars
in
Visv
e
sv
a
ra
y
a
Tec
h
n
o
l
o
g
ica
l
U
n
iv
e
rsit
y
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
r
o
o
p
a
sh
r
e
e
h
r@g
ss
s.e
d
u
.
i
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.