I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
, pp.
3014
~
3021
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
14
.i
4
.pp
3014
-
3021
3014
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
M
ac
h
i
n
e
l
e
ar
n
i
n
g ap
p
l
i
c
at
i
on
f
or
p
ar
t
i
c
l
e
ac
c
e
l
e
r
at
or
o
p
t
i
m
i
z
at
i
on
-
a r
e
vi
e
w
I
s
t
i
D
ia
n
R
ac
h
m
aw
at
i
1,
2
, N
az
r
u
l
E
f
f
e
n
d
y
1
, T
au
f
ik
2
1
I
nt
e
l
l
i
ge
nt
a
nd E
m
be
dde
d S
ys
t
e
m
R
e
s
e
a
r
c
h G
r
oup, D
e
pa
r
t
m
e
nt
of
N
uc
l
e
a
r
E
ngi
ne
e
r
i
ng a
nd E
ngi
ne
e
r
i
ng
P
hys
i
c
s
,
F
a
c
ul
t
y of
E
ngi
ne
e
r
i
ng, U
ni
ve
r
s
i
t
a
s
G
a
dj
a
h M
a
da
, Y
ogya
k
a
r
t
a
, I
ndone
s
i
a
2
R
e
s
e
a
r
c
h
C
e
nt
e
r
f
or
A
c
c
e
l
e
r
a
t
or
T
e
c
hnol
ogy, R
e
s
e
a
r
c
h O
r
ga
ni
z
a
t
i
on f
or
N
uc
l
e
a
r
E
ne
r
gy, N
a
t
i
ona
l
R
e
s
e
a
r
c
h a
nd I
nnova
t
i
on A
ge
nc
y
,
S
out
h T
a
nge
r
a
ng
, I
ndone
s
i
a
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
O
c
t
25
,
2024
R
e
vi
s
e
d
J
un
12
,
2025
A
c
c
e
pt
e
d
J
ul
10
,
2025
Partic
le
acce
lerat
ors
rece
ive
signific
ant
attenti
on
from
resea
rche
rs
.
This
machine
consists
of
various
interdepe
ndent
elements,
so
it
is
complex.
Efficient
system
tuning
and
diagnostics
are
essential
for
utilizing
acce
lerator
technology.
In
addition,
machine
learning
(ML)
has
been
applied
in
several
applicati
ons.
ML
methods
such
as
artificia
l
neural
networks,
random
forest,
reinforcement
learning,
genetic
algorithm
,
and
Bayesian
optimizat
io
n
have
been
used
for
accelerator
optimization.
The
optimization
of
p
article
accelerators
covers
their
performance
and
efficiency.
This
paper
revie
ws
the
applicati
on
of
ML
techniques
in
optimizing
particle
accele
rators,
highlighting
their
importance
in
addressing
the
complexity
inher
ent
in
accelerator s
ystems
and advanci
ng accelerato
r science an
d technol
ogy.
K
e
y
w
o
r
d
s
:
A
c
c
e
le
r
a
to
r
M
a
c
hi
ne
l
e
a
r
ni
ng
N
e
ur
a
l
ne
twor
ks
O
pt
im
iz
a
ti
on
P
a
r
ti
c
le
R
a
ndom f
or
e
s
t
This is an
open
acce
ss artic
le unde
r the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
N
a
z
r
ul
E
f
f
e
ndy
I
n
t
e
ll
i
g
e
nt
a
n
d
E
m
b
e
d
d
e
d
S
y
s
t
e
m
R
e
s
e
a
r
c
h
G
r
o
u
p
,
D
e
p
a
r
tm
e
n
t
o
f
N
u
c
l
e
a
r
E
n
gi
n
e
e
r
in
g
a
n
d
E
ng
i
n
e
e
r
i
n
g
P
h
y
s
i
c
s
F
a
c
ul
ty
of
E
ngi
ne
e
r
in
g, U
ni
ve
r
s
it
a
s
G
a
dj
a
h M
a
da
S
t.
G
r
a
f
ik
a
2, Y
ogya
ka
r
ta
, I
ndone
s
ia
E
m
a
il
:
na
z
r
ul
@
ugm
.a
c
.i
d
1.
I
N
T
R
O
D
U
C
T
I
O
N
P
a
r
ti
c
l
e
a
c
c
e
l
e
r
a
to
r
s
a
c
c
e
l
e
r
a
te
c
ha
r
ge
d
p
a
r
t
ic
le
s
a
t
a
t
om
i
c
a
n
d
s
ub
a
t
om
i
c
s
i
z
e
s
[
1]
.
P
a
r
ti
c
l
e
a
c
c
e
l
e
r
a
to
r
s
pl
a
y
c
r
u
c
i
a
l
r
ol
e
in
in
du
s
tr
ia
l
a
p
pl
i
c
a
ti
on
s
,
s
c
ie
nt
if
ic
r
e
s
e
a
r
c
h,
a
nd
h
e
a
lt
h
c
a
r
e
,
in
c
lu
di
n
g
pr
o
du
c
ti
on
of
r
a
d
io
i
s
ot
o
pe
s
[
2]
,
nu
c
l
e
a
r
f
or
e
n
s
i
c
s
[
3]
,
g
e
n
e
ti
c
m
ut
a
ti
o
n
[
4]
,
[
5]
,
a
c
c
e
le
r
a
t
or
-
dr
iv
e
n
s
y
s
t
e
m
s
[
6]
–
[
8]
,
nu
c
le
a
r
la
b
or
a
to
r
ie
s
,
m
a
t
e
r
i
a
l
s
r
e
s
e
a
r
c
h
[
9]
–
[
12]
,
a
nd
b
or
on
ne
ut
r
o
n
c
a
p
tu
r
e
t
he
r
a
py.
P
r
ot
o
n
s
a
n
d
e
l
e
c
tr
on
s
,
w
hi
c
h
a
r
e
c
h
a
r
g
e
d
w
it
h
a
to
m
ic
p
a
r
ti
c
l
e
s
,
c
om
pr
i
s
e
m
o
s
t
of
th
e
p
a
r
ti
c
l
e
s
tr
e
a
m
.
G
e
ne
r
a
l
ly
,
pa
r
ti
c
l
e
a
c
c
e
le
r
a
t
or
s
a
r
e
de
v
e
l
op
e
d
a
c
c
or
d
in
g t
o t
he
ir
s
p
e
c
if
i
c
pur
po
s
e
s
,
a
n
d
t
he
ty
pe
of
a
p
pl
i
c
a
ti
o
n d
e
p
e
nd
s
on
a
c
c
e
l
e
r
a
to
r
'
s
e
n
e
r
gy
.
S
om
e
p
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
ha
ve
c
om
pl
e
x
e
xpe
r
im
e
nt
a
l
in
s
ta
ll
a
ti
on
s
a
nd
pr
oduc
e
di
r
e
c
te
d
be
a
m
s
of
hi
gh
-
e
ne
r
gy
pa
r
ti
c
le
s
to
w
a
r
d
ta
r
ge
ts
.
T
he
m
a
in
c
om
pone
nt
s
of
a
n
a
c
c
e
le
r
a
to
r
c
ons
is
t
of
th
e
c
ha
r
ge
d
p
a
r
ti
c
le
be
a
m
s
our
c
e
or
in
je
c
to
r
,
a
c
c
e
l
e
r
a
ti
on
s
ys
te
m
,
va
c
uum
tu
be
s
ys
te
m
,
opt
ic
s
ys
te
m
,
ta
r
ge
t
s
y
s
te
m
,
a
nd
in
s
tr
um
e
nt
a
ti
on
a
nd
c
ont
r
ol
s
ys
te
m
.
T
he
in
te
r
r
e
la
ti
ons
hi
ps
a
m
ong
th
e
s
ys
te
m
s
r
e
s
ul
t
in
hi
gh
c
om
pl
e
xi
ty
.
C
ons
id
e
r
in
g
th
e
c
om
pl
e
xi
ty
of
e
a
c
h
s
ub
s
ys
te
m
a
nd
th
e
unpr
e
di
c
ta
bi
li
ty
of
in
te
r
a
c
ti
ons
a
m
ong
th
e
m
,
it
i
s
pr
e
tt
y
c
ha
ll
e
ngi
ng
to
a
voi
d
f
a
il
ur
e
s
a
nd
ope
r
a
ti
ona
l
e
r
r
or
s
[
13]
.
N
a
vi
ga
ti
ng
th
e
nonl
in
e
a
r
f
unc
ti
ons
of
th
e
c
om
pone
nt
s
a
nd
dyna
m
ic
m
a
c
hi
ne
s
e
tt
in
gs
in
a
c
c
e
l
e
r
a
to
r
opt
im
iz
a
ti
on
is
a
s
ig
ni
f
ic
a
nt
c
ha
ll
e
nge
a
f
f
e
c
ti
ng
pa
r
ti
c
le
be
a
m
de
s
ig
n, ope
r
a
ti
on, a
nd c
ont
r
ol
[
14]
.
P
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
a
r
e
nonl
in
e
a
r
s
y
s
te
m
s
,
a
nd
f
ur
th
e
r
r
e
s
e
a
r
c
h
is
ne
c
e
s
s
a
r
y
due
to
th
e
ir
c
om
pl
e
xi
ty
[
15]
.
T
he
r
e
a
r
e
m
a
ny
in
tr
in
s
ic
nonl
in
e
a
r
in
te
r
a
c
ti
ons
be
twe
e
n
it
s
s
ys
te
m
c
om
pone
nt
s
.
I
t
is
c
ha
ll
e
ngi
ng
to
na
vi
ga
te
th
r
ough
th
e
nonl
in
e
a
r
f
unc
ti
ons
of
th
ous
a
nds
of
c
om
pone
nt
s
a
nd
dyna
m
ic
m
a
c
hi
ne
s
e
tt
in
gs
in
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
M
ac
hi
ne
l
e
ar
ni
ng appli
c
at
io
n f
or
par
ti
c
le
ac
c
e
le
r
at
or
opt
imi
z
at
io
n
-
a r
e
v
ie
w
(
I
s
ti
D
ia
n R
ac
hm
aw
at
i
)
3015
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
opt
im
iz
a
ti
on
[
16]
.
T
h
e
s
e
f
a
c
to
r
s
a
f
f
e
c
t
p
a
r
ti
c
le
be
a
m
de
s
ig
n,
ope
r
a
ti
on,
a
nd
c
ont
r
ol
.
C
onve
nt
io
na
l
m
e
th
ods
ha
ve
not
be
e
n
s
uc
c
e
s
s
f
ul
in
th
is
do
m
a
in
,
le
a
di
ng
to
c
ons
ta
nt
a
nd
c
o
s
tl
y
s
y
s
te
m
m
oni
to
r
in
g
by
hu
m
a
n
ope
r
a
to
r
s
.
A
r
ti
f
ic
ia
l
in
te
ll
ig
e
nc
e
(
A
I
)
it
s
e
lf
ha
s
be
e
n
w
id
e
ly
a
ppl
ie
d
in
s
e
ve
r
a
l
a
ppl
ic
a
ti
ons
[
17]
–
[
19]
.
A
I
a
lg
or
it
hm
s
a
r
e
e
s
s
e
nt
ia
l
f
or
c
ont
r
ol
,
tu
ni
ng
[
20]
,
di
a
gnos
ti
c
s
[
21]
,
a
nd
m
ode
li
ng
of
a
c
c
e
le
r
a
to
r
s
. V
a
r
io
us
m
a
c
hi
ne
l
e
a
r
ni
ng (
M
L
)
m
e
th
ods
ha
ve
be
e
n ut
il
iz
e
d f
or
a
c
c
e
le
r
a
to
r
de
ve
lo
pm
e
nt
.
D
e
s
ig
ni
ng
a
c
c
e
le
r
a
to
r
s
m
or
e
e
f
f
ic
ie
nt
ly
m
a
y
be
a
c
c
om
pl
is
he
d
by
ut
il
iz
in
g
M
L
te
c
hni
que
s
.
U
s
in
g
s
ophi
s
ti
c
a
te
d
opt
im
iz
a
ti
on
m
e
th
od
s
a
nd
d
a
ta
-
in
te
ns
iv
e
a
ppr
oa
c
he
s
,
M
L
m
a
y
boos
t
pr
oduc
ti
vi
ty
,
a
c
c
e
le
r
a
te
de
s
ig
n,
a
nd
e
nha
nc
e
th
e
a
c
c
e
le
r
a
to
r
'
s
pe
r
f
or
m
a
nc
e
.
T
he
a
lg
or
it
hm
s
m
ig
ht
e
xa
m
in
e
la
r
ge
da
ta
s
e
ts
f
r
om
pr
e
vi
ous
a
c
c
e
le
r
a
to
r
de
s
ig
n
s
a
nd
s
im
ul
a
ti
ons
to
f
in
d
tr
e
nds
a
n
d
opt
im
iz
e
s
e
tt
in
gs
f
or
de
s
ir
e
d
r
e
s
ul
ts
.
U
s
in
g
m
a
s
s
iv
e
da
ta
s
e
ts
c
ont
a
in
in
g
pa
s
t
pe
r
f
or
m
a
nc
e
a
nd
e
xpe
r
im
e
nt
a
l
out
c
om
e
s
,
r
e
s
e
a
r
c
he
r
s
m
a
y
tr
a
in
M
L
m
ode
ls
to
f
in
d
pa
tt
e
r
ns
a
nd
a
s
s
oc
ia
ti
ons
th
a
t
he
lp
gui
de
th
e
de
s
ig
n
a
nd
m
a
na
ge
m
e
nt
of
vi
ta
l
a
c
c
e
le
r
a
to
r
pa
r
ts
.
F
or
in
s
ta
nc
e
,
M
L
a
lg
or
it
hm
s
c
a
n
a
s
s
i
s
t
in
opt
im
iz
in
g
th
e
de
s
ig
n
c
a
vi
ty
'
s
f
or
m
a
nd
m
a
te
r
ia
l
c
om
pos
it
io
n
t
o
in
c
r
e
a
s
e
pa
r
ti
c
le
a
c
c
e
le
r
a
ti
on e
f
f
ic
ie
nc
y.
M
or
e
ove
r
,
M
L
c
a
n
s
uppor
t
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
ys
te
m
s
'
s
ta
bi
li
ty
a
nd
c
ont
r
ol
.
M
L
a
lg
or
it
hm
s
c
a
n
e
nha
nc
e
t
he
c
ont
r
ol
s
e
tt
in
gs
f
or
s
tr
e
ngt
h a
nd pe
r
f
or
m
a
nc
e
, r
e
s
ul
ti
ng i
n m
or
e
e
f
f
ic
ie
nt
ope
r
a
ti
on, by e
va
lu
a
ti
ng
r
e
a
l
-
ti
m
e
s
e
ns
or
da
ta
a
nd
us
in
g
pr
e
di
c
ti
ve
m
ode
li
ng.
T
he
te
c
h
ni
que
s
f
ol
lo
w
th
e
goa
ls
to
be
a
c
hi
e
ve
d.
T
hi
s
pa
pe
r
r
e
vi
e
w
s
va
r
io
us
M
L
te
c
hni
qu
e
s
a
nd
a
ppl
ic
a
ti
ons
f
or
a
c
c
e
l
e
r
a
to
r
s
.
B
y
c
onduc
ti
ng
a
r
e
vi
e
w
,
it
i
s
e
xpe
c
te
d
th
a
t
knowle
dge
w
il
l
be
obt
a
in
e
d,
na
m
e
ly
knowing
w
ha
t
te
c
hni
que
s
e
xi
s
t
in
M
L
,
gr
oupi
ng
M
L
m
e
th
ods
ba
s
e
d
on
pr
obl
e
m
s
f
a
c
e
d
in
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
,
th
e
a
dva
nt
a
ge
s
of
th
e
s
e
m
e
th
ods
,
a
nd
th
e
r
e
qui
r
e
m
e
nt
s
t
ha
t
m
us
t
be
m
e
t
to
opt
im
i
z
e
us
in
g M
L
.
2.
M
E
T
H
O
D
T
he
r
e
s
e
a
r
c
h
que
s
ti
ons
f
or
M
L
in
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
r
e
vol
ve
a
r
ound
opt
im
iz
in
g
pa
r
a
m
e
te
r
s
,
id
e
nt
if
yi
ng
ut
il
iz
e
d
M
L
m
e
th
ods
,
a
nd
unde
r
s
ta
ndi
ng
tr
e
nds
i
n
a
c
c
e
l
e
r
a
to
r
M
L
a
ppl
ic
a
ti
ons
.
K
e
yw
or
d
a
nd
li
te
r
a
tu
r
e
s
e
a
r
c
h i
s
vi
ta
l
f
or
i
de
nt
if
yi
ng
r
e
le
va
nt
l
it
e
r
a
tu
r
e
t
hr
ou
gh a
ppr
opr
ia
te
ke
yw
or
ds
a
nd s
e
a
r
c
h s
tr
a
te
gi
e
s
,
u
s
in
g
B
ool
e
a
n
ope
r
a
to
r
s
to
r
e
f
in
e
s
e
a
r
c
he
s
.
W
e
r
e
vi
e
w
r
e
tr
ie
ve
d
doc
um
e
nt
ti
tl
e
s
a
nd
a
bs
tr
a
c
t
s
to
a
s
s
e
s
s
r
e
le
va
nc
e
to
th
e
r
e
s
e
a
r
c
h
que
s
ti
on,
doc
um
e
nt
in
g
th
e
s
e
a
r
c
h
pr
oc
e
s
s
m
e
ti
c
ul
ou
s
ly
f
or
tr
a
ns
pa
r
e
nc
y
a
nd
r
e
pr
oduc
ib
il
it
y.
K
now
le
dge
e
xt
r
a
c
ti
on
in
vol
ve
s
s
ynt
he
s
iz
in
g
pe
r
ti
ne
nt
in
s
ig
ht
s
f
r
om
va
r
io
us
s
our
c
e
s
to
a
ddr
e
s
s
r
e
s
e
a
r
c
h
obj
e
c
ti
ve
s
a
nd
or
ga
ni
z
in
g
a
nd
in
te
r
pr
e
ti
ng
in
f
or
m
a
ti
on
s
ys
te
m
a
ti
c
a
ll
y
to
de
r
iv
e
m
e
a
ni
ngf
ul
in
s
ig
ht
s
.
C
r
it
ic
a
l
e
v
a
lu
a
ti
on
e
ns
ur
e
s
th
e
in
te
gr
it
y
of
e
xt
r
a
c
te
d
k
now
le
dge
,
f
a
c
il
it
a
ti
ng
s
ub
s
e
que
nt
a
n
a
ly
s
is
a
nd
in
te
r
pr
e
ta
ti
on.
K
now
le
dge
di
f
f
e
r
e
nt
ia
ti
on
c
a
te
gor
iz
e
s
a
nd
or
ga
ni
z
e
s
e
xt
r
a
c
te
d
knowle
dge
b
a
s
e
d
on
th
e
m
e
s
,
pa
tt
e
r
ns
, or
va
r
ia
ti
ons
, de
e
pe
ni
ng unde
r
s
ta
ndi
ng
a
nd e
na
bl
in
g
m
or
e
e
f
f
e
c
ti
ve
a
na
ly
s
is
.
F
ig
ur
e
1
s
how
s
th
e
a
n
a
ly
s
is
u
s
in
g
V
O
S
vi
e
w
e
r
.
T
he
r
e
is
a
s
tr
ong
c
onne
c
ti
on
be
twe
e
n
th
e
f
ie
ld
s
of
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
te
c
hnol
ogy
a
nd
AI
a
na
ly
s
is
us
in
g
V
O
S
v
ie
w
e
r
.
T
hi
s
is
be
c
a
us
e
ne
ur
a
l
ne
twor
ks
a
nd
ot
he
r
A
I
te
c
hni
que
s
a
r
e
in
c
r
e
a
s
in
gl
y
us
e
d
to
c
ont
r
ol
a
nd
opt
im
iz
e
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
.
T
he
us
e
of
ne
ur
a
l
ne
twor
ks
a
nd othe
r
A
I
m
e
th
ods
i
n
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
t
e
c
hnol
ogy is
a
r
a
p
id
ly
gr
ow
in
g
f
ie
ld
. A
s
A
I
t
e
c
hni
que
s
c
ont
in
ue
to
de
ve
lo
p, w
e
c
a
n e
xp
e
c
t
to
s
e
e
e
ve
n m
or
e
i
nnova
ti
ve
a
ppl
ic
a
ti
ons
i
n t
hi
s
f
ie
ld
.
F
ig
ur
e
1. T
r
e
nds
of
ML
a
ppl
ic
a
ti
on
s
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
opt
im
iz
a
ti
on
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
3014
-
3021
3016
3.
M
A
C
H
I
N
E
L
E
A
R
N
I
N
G
A
L
G
O
R
I
T
H
M
M
L
be
c
om
e
s
a
tt
r
a
c
ti
ve
due
to
th
e
a
bunda
nc
e
of
da
ta
.
I
t
c
a
n
le
a
r
n
c
om
pl
e
x
pa
tt
e
r
ns
,
pr
e
di
c
t
out
c
om
e
s
,
a
nd
a
ut
om
a
te
pr
oc
e
s
s
e
s
.
M
L
ge
ne
r
a
ll
y
le
a
r
ns
r
e
la
ti
o
ns
hi
ps
be
tw
e
e
n
in
put
a
nd
out
put
f
r
om
e
xi
s
ti
ng
da
ta
[
22]
.
T
a
s
ks
pe
r
f
or
m
e
d
by
M
L
in
c
lu
de
r
e
gr
e
s
s
io
n,
c
la
s
s
if
ic
a
ti
on,
c
lu
s
te
r
in
g,
a
nd
a
nom
a
ly
de
te
c
ti
on
pr
obl
e
m
s
[
23]
,
[
24]
.
S
e
ve
r
a
l
M
L
m
e
th
ods
,
of
te
n
c
om
bi
ne
d
w
it
h
opt
im
iz
a
ti
on
te
c
hni
que
s
, a
r
e
e
xpl
a
in
e
d
in
th
e
f
ol
lo
w
in
g s
ub
-
s
e
c
ti
ons
.
3.1.
A
r
t
if
ic
ia
l
n
e
u
r
al
n
e
t
w
or
k
s
O
ne
of
th
e
m
os
t
f
a
m
ous
a
nd
c
om
m
onl
y
us
e
d
M
L
a
lg
or
it
h
m
s
is
a
r
ti
f
ic
ia
l
ne
ur
a
l
ne
twor
ks
(
A
N
N
)
[
25]
.
A
N
N
is
w
e
ll
-
s
ui
te
d
f
o
r
le
a
r
ni
ng
ta
s
ks
w
he
r
e
da
ta
in
c
lu
de
s
noi
s
e
,
c
om
pl
e
x
s
ig
na
ls
,
a
nd
ta
r
ge
t
out
put
f
unc
ti
ons
th
a
t
m
a
y
c
ons
i
s
t
of
m
ul
ti
pl
e
pa
r
a
m
e
te
r
s
.
A
n
e
ur
a
l
ne
twor
k
c
a
n
be
d
e
f
in
e
d
a
s
a
c
ol
le
c
ti
on
of
f
unc
ti
ons
w
it
h
w
e
ig
ht
e
d
c
onne
c
ti
ons
a
m
ong
th
e
m
.
T
he
s
e
w
e
i
ght
e
d
c
onne
c
ti
ons
c
a
n
be
a
dj
u
s
te
d
or
tr
a
in
e
d
th
r
ough
a
n
a
ut
om
a
te
d
opt
im
iz
a
ti
on
pr
oc
e
s
s
unt
il
th
e
d
e
s
ir
e
d
o
ut
put
be
ha
vi
or
is
a
c
hi
e
ve
d.
T
r
a
in
in
g
m
a
y
a
ls
o
in
vol
ve
c
ha
nge
s
to
th
e
s
tr
uc
tu
r
a
l
c
om
pone
nt
s
of
th
e
ne
twor
k,
s
uc
h
a
s
th
e
num
be
r
of
node
s
a
nd
la
ye
r
s
[
26]
.
N
e
ur
a
l
ne
twor
ks
c
a
n
b
e
tr
a
in
e
d
us
in
g
s
im
ul
a
ti
on
da
ta
,
m
e
a
s
ur
a
bl
e
da
ta
,
or
a
c
om
bi
na
ti
on
of
bot
h.
M
a
ny
tr
a
in
in
g a
ppr
oa
c
he
s
a
nd a
r
c
hi
te
c
tu
r
e
s
a
r
e
a
va
il
a
bl
e
,
e
a
c
h
s
ui
ta
bl
e
f
or
s
pe
c
if
ic
pr
obl
e
m
c
la
s
s
e
s
.
A
N
N
is
w
id
e
ly
us
e
d
in
va
r
io
us
f
ie
ld
s
,
f
r
om
s
a
f
e
ty
to
c
r
it
ic
a
l
a
r
e
a
s
s
uc
h
a
s
a
c
c
e
le
r
a
to
r
s
[
27]
.
A
N
N
be
lo
ngs
to
f
a
m
il
ia
r
M
L
m
e
th
ods
th
a
t
a
r
e
f
r
e
que
nt
ly
u
s
e
d
in
a
c
c
e
le
r
a
to
r
a
ppl
ic
a
ti
ons
.
S
e
v
e
r
a
l
s
tu
di
e
s
h
a
ve
be
e
n
c
onduc
te
d
on
th
e
us
e
of
ne
ur
a
l
ne
twor
ks
[
28]
.
A
N
N
,
e
s
pe
c
ia
ll
y
de
e
p
le
a
r
ni
ng
m
ode
ls
,
a
r
e
pow
e
r
f
ul
to
ol
s
f
or
le
a
r
ni
ng
c
om
pl
e
x
pa
tt
e
r
ns
a
nd
r
e
la
ti
ons
hi
ps
f
r
om
da
ta
.
T
he
y
c
a
n
be
us
e
d
f
or
ta
s
ks
s
uc
h
a
s
s
ur
r
oga
te
m
ode
ll
in
g,
w
he
r
e
th
e
ne
ur
a
l
ne
twor
k
le
a
r
ns
to
a
ppr
oxi
m
a
te
th
e
pe
r
f
or
m
a
nc
e
of
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
c
om
pone
nt
s
ba
s
e
d on input
pa
r
a
m
e
te
r
s
. A
N
N
c
a
n a
ls
o be
i
nt
e
g
r
a
te
d i
nt
o opti
m
iz
a
ti
on a
lg
or
it
hm
s
t
o
gui
de
t
he
s
e
a
r
c
h
pr
oc
e
s
s
m
or
e
e
f
f
e
c
ti
ve
ly
.
T
he
a
ppl
ic
a
ti
ons
of
A
N
N
in
c
lu
de
be
a
m
dyna
m
ic
s
opt
im
iz
a
ti
on
[
29]
,
c
ont
r
o
l
[
30]
, s
ur
r
oga
te
m
ode
l
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
[
31]
, pha
s
e
s
p
a
c
e
di
a
gnos
ti
c
s
[
32]
, a
nd
opt
ic
s
r
e
c
on
s
tr
uc
ti
on.
3.2.
R
an
d
om
f
or
e
s
t
R
a
ndom
f
or
e
s
t
is
a
n
a
lg
or
it
hm
th
a
t
c
a
n
be
us
e
d
f
or
r
e
gr
e
s
s
io
n
a
nd
c
la
s
s
if
ic
a
ti
on
a
na
ly
s
is
.
R
a
ndom
f
or
e
s
t
is
a
ve
r
s
a
ti
le
M
L
m
e
th
od
th
a
t
c
a
n
be
e
f
f
e
c
ti
ve
ly
a
ppl
ie
d
to
va
r
io
us
a
s
pe
c
ts
of
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
opt
im
iz
a
ti
on,
in
c
lu
di
ng
s
ur
r
oga
te
m
ode
ll
in
g,
f
e
a
tu
r
e
im
por
ta
n
c
e
a
na
ly
s
i
s
,
a
nom
a
ly
de
te
c
ti
on,
a
nd
e
ns
e
m
bl
e
opt
im
iz
a
ti
on.
T
he
r
a
ndom
f
or
e
s
t
m
e
th
od
is
e
f
f
e
c
ti
ve
f
or
in
s
tr
um
e
nt
a
ti
on
e
r
r
or
de
te
c
ti
on,
f
or
e
xa
m
pl
e
,
f
or
id
e
nt
if
yi
ng a
nd c
or
r
e
c
ti
ng e
r
r
or
s
i
n m
a
gne
ts
.
3.3.
R
e
in
f
or
c
e
m
e
n
t
l
e
ar
n
in
g
R
e
in
f
or
c
e
m
e
nt
l
e
a
r
ni
ng (
R
L
)
i
s
a
f
r
a
m
e
w
or
k
i
n w
hi
c
h a
r
t
if
ic
ia
l
a
ge
nt
s
l
e
a
r
n by int
e
r
a
c
ti
ng w
it
h
t
he
ir
e
nvi
r
onm
e
nt
.
R
L
c
a
n
be
us
e
d
to
de
ve
lo
p
s
ur
r
oga
te
m
ode
ls
th
a
t
r
e
pr
oduc
e
r
e
a
l
-
w
or
ld
s
ys
te
m
s
'
be
ha
vi
or
s
a
nd
tr
a
in
onl
in
e
a
ge
nt
s
to
ta
ke
c
ont
r
ol
a
c
ti
ons
in
th
o
s
e
s
ys
te
m
s
[
33]
.
T
he
s
e
onl
in
e
a
g
e
nt
s
w
il
l
ul
ti
m
a
te
ly
c
ont
r
ol
th
e
a
c
tu
a
l
a
c
c
e
l
e
r
a
to
r
s
ys
te
m
.
R
L
ha
s
be
e
n
a
ppl
ie
d
in
c
ont
r
ol
,
or
bi
t
c
or
r
e
c
ti
on
[
34]
,
a
nd
r
e
a
l
-
ti
m
e
f
e
e
dba
c
k
c
ont
r
ol
l
oop
[
35]
.
3.4.
G
e
n
e
t
ic
a
lg
or
it
h
m
G
e
ne
ti
c
a
lg
or
it
hm
(
G
A
)
is
a
n
e
vol
ut
io
na
r
y
opt
im
iz
a
ti
on
te
c
h
ni
que
in
s
pi
r
e
d
by
na
tu
r
a
l
s
e
le
c
ti
on.
T
he
y
a
r
e
w
e
ll
-
s
ui
te
d
f
or
pr
ob
le
m
s
w
it
h
a
n
a
m
pl
e
s
e
a
r
c
h
s
pa
c
e
a
nd
c
om
pl
e
x,
nonl
in
e
a
r
r
e
la
ti
ons
hi
ps
.
G
A
c
a
n
e
f
f
ic
ie
nt
ly
e
xpl
or
e
th
e
de
s
ig
n
s
pa
c
e
of
pa
r
t
ic
le
a
c
c
e
le
r
a
to
r
s
a
nd
id
e
nt
if
y
opt
im
a
l
c
onf
ig
ur
a
ti
ons
f
or
c
om
pone
nt
s
s
uc
h
a
s
c
a
vi
ti
e
s
, m
a
gne
t
s
, a
nd r
a
di
o f
r
e
que
nc
y (
R
F
)
s
ys
te
m
s
.
3.5.
B
aye
s
ia
n
o
p
t
im
iz
at
io
n
B
a
ye
s
ia
n
opt
im
iz
a
ti
on
is
a
pr
oba
bi
li
s
ti
c
opt
im
iz
a
ti
on
te
c
hni
que
th
a
t
us
e
s
s
ur
r
oga
te
m
ode
ls
to
a
ppr
oxi
m
a
te
th
e
obj
e
c
ti
ve
f
unc
ti
on
[
36]
,
[
37]
.
I
t
e
f
f
ic
ie
nt
ly
ba
l
a
nc
e
s
e
xpl
or
a
ti
on
a
nd
e
xpl
oi
ta
ti
on
to
f
in
d
th
e
gl
oba
l
opt
im
um
w
hi
le
m
in
im
iz
in
g
th
e
num
be
r
o
f
e
va
lu
a
ti
ons
.
B
a
ye
s
ia
n
opt
im
iz
a
ti
on
is
e
f
f
e
c
ti
ve
f
or
opt
im
iz
in
g
bl
a
c
k
-
box
f
unc
ti
ons
,
m
a
ki
ng
it
s
ui
ta
bl
e
f
or
opt
im
iz
i
ng
c
om
pl
e
x
s
im
ul
a
ti
ons
of
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
ys
te
m
s
.
4.
M
A
C
H
I
N
E
L
E
A
R
N
I
N
G
I
M
P
L
E
M
E
N
T
A
T
I
O
N
F
O
R
P
A
R
T
I
C
L
E
A
C
C
E
L
E
R
A
T
O
R
P
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
a
r
e
ne
c
e
s
s
a
r
y
f
or
m
a
ny
s
c
ie
nt
if
ic
pr
oj
e
c
ts
,
but
opt
im
iz
in
g
th
e
ir
pe
r
f
o
r
m
a
nc
e
a
nd
r
e
li
a
bi
li
ty
pr
e
s
e
nt
s
s
ig
ni
f
ic
a
nt
c
ha
ll
e
ng
e
s
.
M
L
te
c
hni
qu
e
s
of
f
e
r
pr
om
is
in
g
s
ol
ut
io
ns
f
or
e
nha
n
c
in
g
de
s
ig
n
c
om
pone
nt
s
,
pa
r
a
m
e
te
r
opt
im
iz
a
ti
on,
c
ont
r
ol
,
di
a
gn
os
ti
c
s
,
a
nd
m
od
e
ll
in
g
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
.
P
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
be
n
e
f
it
s
ig
ni
f
ic
a
nt
ly
f
r
om
a
ppl
yi
ng
M
L
te
c
hni
que
s
,
of
f
e
r
in
g
pr
om
is
in
g
s
ol
ut
io
ns
to
be
tt
e
r
de
s
ig
n
c
om
pone
nt
s
,
pr
e
di
c
ti
on,
a
nom
a
ly
de
te
c
ti
on,
p
a
r
a
m
e
te
r
tu
ni
ng,
r
e
a
l
-
ti
m
e
a
da
pt
iv
e
c
ont
r
ol
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
M
ac
hi
ne
l
e
ar
ni
ng appli
c
at
io
n f
or
par
ti
c
le
ac
c
e
le
r
at
or
opt
imi
z
at
io
n
-
a r
e
v
ie
w
(
I
s
ti
D
ia
n R
ac
hm
aw
at
i
)
3017
a
nd
be
a
m
dyna
m
ic
s
.
F
ig
ur
e
2
s
how
s
s
e
ve
r
a
l
M
L
m
e
th
ods
a
nd
th
e
ir
a
ppl
ic
a
ti
ons
f
or
pa
r
ti
c
le
a
c
c
e
l
e
r
a
to
r
opt
im
iz
a
ti
on.
F
ig
ur
e
2.
ML
m
e
th
ods
a
nd t
he
ir
a
ppl
ic
a
ti
ons
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
opt
im
iz
a
ti
on
4.1. De
s
ig
n
c
om
p
on
e
n
t
s
E
ns
ur
in
g
a
de
qua
te
pa
r
ti
c
le
a
c
c
e
le
r
a
ti
on
ne
c
e
s
s
it
a
te
s
opt
im
iz
in
g
a
c
c
e
le
r
a
to
r
c
om
pone
nt
s
,
in
c
lu
di
ng
th
e
de
te
c
to
r
,
hi
gh
-
vol
ta
ge
pul
s
e
tr
a
ns
f
or
m
e
r
s
,
m
a
gne
ti
c
c
om
p
one
nt
s
[
38]
,
R
F
,
c
a
vi
ty
,
a
c
c
e
le
r
a
ti
on
s
ys
te
m
s
,
a
nd
c
ont
r
ol
m
oni
to
r
in
g
s
ys
t
e
m
s
.
M
L
m
ode
ls
c
a
n
e
xa
m
in
e
pa
s
t
da
ta
,
s
im
ul
a
ti
ons
,
a
nd
e
xpe
r
im
e
nt
a
l
out
c
om
e
s
to
de
te
r
m
in
e
th
e
be
s
t
de
s
ig
ns
f
or
th
e
c
om
pone
nt
s
.
U
ti
li
z
in
g
M
L
te
c
hni
que
s
to
de
s
ig
n
th
e
c
om
pone
nt
s
a
nd
de
ve
lo
p
s
ur
r
oga
te
m
ode
ls
is
e
xpe
c
te
d
to
im
pr
ove
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
e
f
f
ic
ie
nc
y,
pe
r
f
or
m
a
nc
e
,
a
nd
r
e
li
a
bi
li
ty
.
4.2. P
r
e
d
ic
t
io
n
an
d
an
om
al
y d
e
t
e
c
t
io
n
T
he
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
be
ha
vi
or
c
a
n
be
pr
e
di
c
te
d
us
in
g
M
L
m
e
th
ods
.
T
hi
s
in
c
lu
de
s
th
e
pr
e
di
c
ti
on
of
lo
w
-
e
ne
r
gy
be
a
m
tr
a
ns
por
t
tu
ni
ng,
ti
m
e
s
e
r
ie
s
f
o
r
e
c
a
s
ti
ng
us
in
g
c
la
s
s
if
ic
a
ti
on
a
ppr
oa
c
he
s
[
39]
,
be
a
m
l
os
s
m
ode
ll
in
g
[
40]
, pr
e
di
c
ti
on of
l
ow
-
e
ne
r
gy be
a
m
t
r
a
ns
por
t
tu
ni
ng, a
nd l
ongi
tu
di
na
l
pha
s
e
s
pa
c
e
[
41]
.
M
L
c
a
n a
ut
om
a
te
a
nd e
xpe
di
te
di
a
gnos
ti
c
pr
oc
e
s
s
e
s
, pr
oduc
in
g
m
or
e
r
e
li
a
bl
e
, hi
gh
-
pe
r
f
or
m
a
nc
e
a
c
c
e
le
r
a
to
r
s
.
O
th
e
r
M
L
a
ppl
ic
a
ti
ons
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
di
a
gnos
ti
c
s
in
c
lu
de
a
nom
a
ly
de
te
c
ti
on
[
42]
.
A
nom
a
ly
de
te
c
ti
on
te
c
hni
que
s
ha
v
e
a
ls
o
be
e
n
a
ppl
ie
d
to
c
le
a
n
m
e
a
s
ur
e
d
da
ta
by
c
om
pa
r
in
g
it
w
it
h
c
lu
s
te
r
in
g
te
c
hni
que
s
.
4.3. P
ar
am
e
t
e
r
t
u
n
in
g
M
L
a
lg
or
it
hm
s
c
a
n
f
a
c
il
it
a
te
r
e
a
l
-
ti
m
e
opt
im
iz
a
ti
on
of
pa
r
a
m
e
te
r
c
ont
r
ol
s
tr
a
te
gi
e
s
ba
s
e
d
on
da
ta
-
dr
iv
e
n
in
s
ig
ht
s
.
B
y
a
na
ly
z
in
g
la
r
ge
da
ta
s
e
ts
of
ope
r
a
ti
ona
l
pa
r
a
m
e
te
r
s
a
nd
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
M
L
m
ode
ls
c
a
n
id
e
nt
if
y
c
or
r
e
la
ti
ons
,
pa
tt
e
r
ns
,
a
nd
opt
im
a
l
c
ont
r
o
l
s
t
r
a
te
gi
e
s
f
or
io
n
s
our
c
e
s
a
nd
ot
he
r
c
r
it
ic
a
l
c
om
pone
nt
s
.
T
hi
s
e
na
bl
e
s
a
da
pt
iv
e
c
ont
r
ol
m
e
c
ha
ni
s
m
s
th
a
t
dyna
m
ic
a
ll
y
a
dj
us
t
ope
r
a
ti
ona
l
pa
r
a
m
e
te
r
s
to
opt
im
iz
e
a
c
c
e
le
r
a
to
r
pe
r
f
or
m
a
nc
e
unde
r
va
r
io
us
c
ondi
ti
ons
.
B
e
a
m
pa
r
a
m
e
te
r
opt
im
iz
a
ti
on
us
e
s
la
s
s
o
r
e
gr
e
s
s
io
n
f
or
onl
in
e
tu
ne
c
or
r
e
c
ti
on
a
nd
ne
ur
a
l
ne
twor
ks
f
or
be
ta
f
unc
ti
on
s
im
ul
a
ti
on
c
or
r
e
c
ti
on
[
43]
.
D
e
te
c
ti
on
of
m
a
gne
ti
c
f
ie
ld
e
r
r
o
r
s
us
in
g
a
ut
oe
nc
ode
r
ne
ur
a
l
ne
twor
ks
,
li
ne
a
r
r
e
gr
e
s
s
io
n,
a
nd
tu
ne
d
f
e
e
dba
c
k
s
to
r
a
ge
r
in
gs
[
44]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
3014
-
3021
3018
4.4.
P
ar
t
ic
le
ac
c
e
le
r
at
or
c
on
t
r
ol
an
d
d
ia
gn
os
t
ic
s
M
L
in
th
e
c
ont
r
ol
o
f
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
c
a
n
be
ut
il
iz
e
d
f
or
s
ys
te
m
f
a
il
ur
e
pr
e
di
c
ti
on,
a
nom
a
ly
de
te
c
ti
on,
c
ont
r
ol
opt
im
iz
a
ti
on,
a
nd
a
ut
om
a
ti
c
c
ont
r
ol
.
S
om
e
a
ppl
ic
a
ti
ons
of
M
L
in
a
c
c
e
le
r
a
to
r
c
ont
r
ol
in
c
lu
de
de
te
c
to
r
c
ont
r
ol
a
nd
c
a
li
br
a
ti
on
[
45]
,
a
ut
om
a
ti
c
be
a
m
pos
it
io
n
c
ont
r
ol
[
46]
,
p
r
e
di
c
ti
ve
a
c
c
e
le
r
a
to
r
c
ont
r
ol
,
be
a
m
m
a
tc
hi
ng
c
ont
r
ol
,
a
da
pt
iv
e
c
ont
r
ol
f
or
be
a
m
di
a
gnos
ti
c
s
[
47]
,
e
le
c
tr
on
bunc
h
pr
of
il
e
de
te
c
ti
on,
a
nd
be
a
m
dyna
m
ic
c
ont
r
ol
[
48]
.
P
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
di
a
gnos
ti
c
s
is
a
c
om
pl
e
x
a
nd
ti
m
e
-
c
ons
um
in
g
pr
oc
e
s
s
th
a
t
id
e
nt
if
ie
s
a
nd
a
ddr
e
s
s
e
s
is
s
ue
s
in
th
e
a
c
c
e
l
e
r
a
to
r
.
M
L
c
a
n
h
e
lp
a
ut
om
a
te
a
nd
e
xpe
di
te
di
a
gno
s
ti
c
pr
oc
e
s
s
e
s
,
r
e
s
ul
ti
ng
in
m
or
e
r
e
li
a
bl
e
a
nd
hi
gh
-
pe
r
f
or
m
a
nc
e
a
c
c
e
le
r
a
to
r
s
.
S
om
e
M
L
a
ppl
ic
a
ti
ons
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
di
a
gno
s
ti
c
s
in
c
lu
de
m
ul
ti
va
r
ia
bl
e
di
a
gno
s
ti
c
s
a
n
d
vi
r
tu
a
l
di
a
gnos
ti
c
s
of
be
a
m
lo
ngi
tu
di
na
l
pr
ope
r
ti
e
s
[
49]
, [
50]
.
4.5.
M
od
e
l
li
n
g
P
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
m
ode
l
li
ng
is
th
e
pr
oc
e
s
s
of
s
im
ul
a
ti
ng
pa
r
ti
c
le
be
ha
vi
o
r
w
it
hi
n
th
e
a
c
c
e
le
r
a
to
r
.
T
hi
s
pr
oc
e
s
s
is
e
s
s
e
nt
ia
l
f
or
de
s
ig
ni
ng,
opt
im
iz
in
g,
a
nd
c
om
m
is
s
io
ni
ng
a
c
c
e
le
r
a
to
r
s
.
M
L
c
a
n
he
lp
im
pr
ove
th
e
a
c
c
ur
a
c
y
a
nd
e
f
f
ic
ie
nc
y
of
p
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
m
ode
l
li
ng.
H
e
r
e
a
r
e
s
om
e
M
L
a
ppl
ic
a
ti
ons
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
m
ode
l
li
ng:
p
r
e
di
c
ti
on
of
lo
w
-
e
ne
r
gy
be
a
m
tr
a
ns
por
t
tu
ni
ng,
ti
m
e
s
e
r
ie
s
f
or
e
c
a
s
ti
ng
us
in
g
c
la
s
s
if
ic
a
ti
on
a
ppr
oa
c
he
s
,
m
od
e
ll
in
g
of
be
a
m
lo
s
s
,
pr
e
di
c
t
io
n
of
lo
w
-
e
ne
r
gy
be
a
m
tr
a
ns
por
t
tu
ni
ng,
unc
e
r
ta
in
ty
a
na
ly
s
is
,
be
a
m
dyn
a
m
ic
s
[
51]
,
[
52]
,
de
ve
lo
pm
e
nt
of
ot
he
r
a
ppl
ic
a
ti
ons
.
T
he
s
tu
dy
of
pa
r
ti
c
le
be
a
m
m
ot
io
n
in
a
c
c
e
le
r
a
to
r
s
c
ove
r
s
pa
r
ti
c
le
in
te
r
a
c
ti
ons
,
e
le
c
tr
om
a
gne
ti
c
f
ie
ld
s
,
a
nd
ot
he
r
e
le
m
e
nt
s
.
M
L
m
e
th
ods
c
a
n m
ode
l,
pr
e
di
c
t,
a
nd opti
m
iz
e
pa
r
ti
c
le
b
e
a
m
be
ha
vi
o
r
.
5.
C
O
N
C
L
U
S
I
O
N
M
L
pr
e
s
e
nt
s
a
pow
e
r
f
ul
to
ol
s
e
t
f
or
a
dva
nc
in
g
pa
r
ti
c
le
a
c
c
e
l
e
r
a
to
r
te
c
hnol
ogi
e
s
,
of
f
e
r
in
g
c
ont
r
ol
,
tu
ni
ng,
di
a
gnos
ti
c
s
, a
nd
m
ode
l
li
ng
im
pr
ove
m
e
nt
s
. T
he
de
s
ig
n a
nd
a
na
ly
s
is
of
a
c
c
e
le
r
a
to
r
be
a
m
dyn
a
m
ic
s
c
a
n
us
e
a
GA
,
pr
e
di
c
ti
on,
a
nd
a
nom
a
ly
de
te
c
ti
on
u
s
in
g
ne
ur
a
l
ne
t
w
or
ks
a
nd
r
a
ndom
f
or
e
s
ts
.
I
n
a
ddi
ti
on,
li
ne
a
r
a
nd
nonl
in
e
a
r
r
e
gr
e
s
s
io
n
c
a
n
he
lp
a
na
ly
z
e
s
ys
te
m
pa
r
a
m
e
te
r
s
,
a
nd
pa
r
a
m
e
te
r
tu
ni
ng
c
a
n
us
e
B
a
ye
s
ia
n
opt
im
iz
a
ti
on
a
nd
c
ont
r
ol
us
in
g
RL
.
T
he
c
om
bi
na
ti
on
of
th
e
s
e
te
c
hni
que
s
a
ll
ow
s
f
or
m
or
e
s
ophi
s
ti
c
a
te
d
opt
im
iz
a
ti
on
a
nd
r
e
s
pons
iv
e
ne
s
s
to
c
ha
ngi
ng
ope
r
a
ti
ona
l
c
ondi
ti
ons
,
im
pr
ovi
ng
th
e
ove
r
a
ll
e
f
f
ic
ie
nc
y
a
nd
pe
r
f
or
m
a
nc
e
of
th
e
a
c
c
e
le
r
a
to
r
.
S
e
ve
r
a
l
r
e
qui
r
e
m
e
nt
s
m
us
t
be
a
ddr
e
s
s
e
d
to
im
pl
e
m
e
nt
M
L
-
ba
s
e
d
opt
im
iz
a
ti
on
f
or
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
.
F
ir
s
tl
y,
h
ig
h
-
qua
li
ty
a
nd
r
e
pr
e
s
e
nt
a
ti
ve
da
ta
s
e
ts
a
r
e
e
s
s
e
nt
ia
l
f
or
tr
a
in
in
g
a
c
c
ur
a
te
M
L
m
ode
ls
.
T
he
da
ta
s
e
ts
s
houl
d
e
nc
o
m
pa
s
s
va
r
io
us
ope
r
a
ti
ona
l
c
ondi
ti
ons
a
nd
pe
r
f
or
m
a
nc
e
m
e
tr
ic
s
,
e
ns
ur
in
g
r
obus
t
m
ode
l
tr
a
in
in
g
a
nd
va
li
da
ti
on.
A
ddi
ti
ona
ll
y,
c
ol
la
bor
a
ti
on
be
twe
e
n
dom
a
in
e
xpe
r
ts
,
da
ta
s
c
ie
nt
is
ts
,
a
nd
M
L
s
pe
c
ia
li
s
ts
is
ne
c
e
s
s
a
r
y
to
de
ve
lo
p
e
f
f
e
c
ti
ve
opt
im
iz
a
ti
on
s
tr
a
te
gi
e
s
th
a
t
a
ddr
e
s
s
th
e
uni
que
c
ha
ll
e
nge
s
of
p
a
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
y
s
te
m
s
.
C
ont
in
ue
d
r
e
s
e
a
r
c
h
a
nd
d
e
ve
lo
pm
e
nt
in
M
L
a
ppl
ic
a
ti
ons
pr
om
is
e
s
to
e
nha
nc
e
pa
r
ti
c
le
a
c
c
e
le
r
a
to
r
s
'
pe
r
f
or
m
a
nc
e
a
nd
r
e
li
a
bi
li
ty
f
ur
th
e
r
,
dr
iv
in
g
s
c
ie
nt
if
ic
di
s
c
ove
r
y a
nd i
nnova
ti
on.
A
C
K
N
O
WL
E
D
G
M
E
N
T
S
T
he
a
ut
hor
s
th
a
nk
th
e
D
ir
e
c
to
r
a
te
of
T
a
le
nt
M
a
na
ge
m
e
nt
of
t
he
N
a
ti
ona
l
R
e
s
e
a
r
c
h
a
nd
I
nnova
ti
on
A
ge
nc
y,
th
e
R
e
s
e
a
r
c
h
C
e
nt
e
r
f
or
A
c
c
e
le
r
a
to
r
T
e
c
hnol
ogy,
a
nd
U
ni
ve
r
s
it
a
s
G
a
dj
a
h
M
a
da
f
or
th
e
f
a
c
il
it
y
s
uppor
t
f
or
t
hi
s
r
e
s
e
a
r
c
h.
F
U
N
D
I
N
G
I
N
F
O
R
M
A
T
I
O
N
T
hi
s
r
e
s
e
a
r
c
h
w
a
s
f
unde
d
by
th
e
D
i
r
e
c
to
r
a
te
of
R
e
s
e
a
r
c
h,
T
e
c
hnol
ogy,
a
nd
C
om
m
uni
ty
S
e
r
vi
c
e
,
M
in
is
tr
y of
E
duc
a
ti
on, C
ul
tu
r
e
, R
e
s
e
a
r
c
h, a
nd T
e
c
hnol
ogy of
t
he
R
e
publ
ic
of
I
ndone
s
ia
, t
hr
ough the
m
a
s
te
r
’
s
th
e
s
is
r
e
s
e
a
r
c
h
s
c
h
e
m
e
,
unde
r
gr
a
nt
num
be
r
s
048/
E
5/
P
G
.02.00.P
L
/2
024
a
nd
2886/UN1/DI
T
L
I
T
/P
T
.01.03/2024.
A
U
T
H
O
R
C
O
N
T
R
I
B
U
T
I
O
N
S
S
T
A
T
E
M
E
N
T
T
hi
s
jo
ur
na
l
u
s
e
s
th
e
C
ont
r
ib
ut
or
R
ol
e
s
T
a
xonomy
(
C
R
e
di
T
)
to
r
e
c
ogni
z
e
in
di
vi
dua
l
a
ut
hor
c
ont
r
ib
ut
io
ns
, r
e
duc
e
a
ut
hor
s
hi
p di
s
put
e
s
,
a
nd f
a
c
il
it
a
te
c
ol
la
bo
r
a
ti
on.
N
am
e
o
f
A
u
t
h
or
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
I
s
ti
D
ia
n R
a
c
hm
a
w
a
ti
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
N
a
z
r
ul
E
f
f
e
ndy
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
T
a
uf
ik
✓
✓
✓
✓
✓
✓
✓
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
M
ac
hi
ne
l
e
ar
ni
ng appli
c
at
io
n f
or
par
ti
c
le
ac
c
e
le
r
at
or
opt
imi
z
at
io
n
-
a r
e
v
ie
w
(
I
s
ti
D
ia
n R
ac
hm
aw
at
i
)
3019
C
:
C
onc
e
pt
ua
l
i
z
a
t
i
on
M
:
M
e
t
hodol
ogy
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
da
t
i
on
Fo
:
Fo
r
m
a
l
a
na
l
ys
i
s
I
:
I
nve
s
t
i
ga
t
i
on
R
:
R
e
s
our
c
e
s
D
:
D
a
t
a
C
ur
a
t
i
on
O
:
W
r
i
t
i
ng
-
O
r
i
gi
na
l
D
r
a
f
t
E
:
W
r
i
t
i
ng
-
R
e
vi
e
w
&
E
di
t
i
ng
Vi
:
Vi
s
ua
l
i
z
a
t
i
on
Su
:
Su
pe
r
vi
s
i
on
P
:
P
r
oj
e
c
t
a
dm
i
ni
s
t
r
a
t
i
on
Fu
:
Fu
ndi
ng a
c
qui
s
i
t
i
on
C
O
N
F
L
I
C
T
O
F
I
N
T
E
R
E
S
T
S
T
A
T
E
M
E
N
T
T
he
a
ut
hor
s
de
c
la
r
e
th
a
t
th
e
y
h
a
ve
no
kno
w
n
c
om
pe
ti
ng
f
in
a
nc
ia
l
in
te
r
e
s
ts
or
p
e
r
s
ona
l
r
e
la
ti
ons
hi
p
s
th
a
t
c
oul
d ha
ve
a
ppe
a
r
e
d t
o i
nf
lu
e
nc
e
t
h
e
w
or
k
r
e
por
te
d i
n t
hi
s
pa
pe
r
. A
ut
hor
s
s
ta
te
no c
onf
li
c
t
of
i
nt
e
r
e
s
t.
D
A
T
A
A
V
A
I
L
A
B
I
L
I
T
Y
D
a
ta
a
va
il
a
bi
li
ty
is
not
a
ppl
ic
a
bl
e
to
th
is
p
a
pe
r
a
s
no
n
e
w
da
ta
w
e
r
e
c
r
e
a
te
d
or
a
na
ly
z
e
d
in
th
is
s
tu
dy.
R
E
F
E
R
E
N
C
E
S
[
1]
V
i
ka
s
a
nd
R
.
K
.
S
a
hu,
“
A
r
e
vi
e
w
on
a
ppl
i
c
a
t
i
on
of
l
a
s
e
r
t
r
a
c
ke
r
i
n
pr
e
c
i
s
i
on
pos
i
t
i
oni
ng
m
e
t
r
ol
ogy
o
f
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
P
r
e
c
i
s
i
on E
ngi
ne
e
r
i
ng
, vol
. 71, pp. 232
–
249, S
e
p. 2021, doi
:
10.1016/
j
.pr
e
c
i
s
i
o
ne
ng.2021.03.015.
[
2]
J
.
P
.
M
e
i
e
r
e
t
a
l
.
,
“
A
c
c
e
l
e
r
a
t
o
r
-
ba
s
e
d
p
r
odu
c
t
i
on
o
f
S
c
a
n
di
um
r
a
di
oi
s
ot
ope
s
f
o
r
a
ppl
i
c
a
t
i
ons
i
n
p
r
os
t
a
t
e
c
a
n
c
e
r
:
t
o
w
a
r
d
bui
l
d
i
n
g
a
pi
p
e
l
i
ne
f
or
r
a
p
i
d
de
v
e
l
opm
e
n
t
o
f
no
ve
l
t
he
r
a
n
os
t
i
c
s
,”
M
ol
e
c
u
l
e
s
,
v
ol
.
2
8,
no
. 1
6,
A
ug.
2
023
,
do
i
:
10
.3
390
/
m
ol
e
c
u
l
e
s
28
16
604
1.
[
3]
A
.
S
i
m
on,
N
.
P
.
B
a
r
r
a
da
s
,
C
.
J
e
yne
s
,
a
nd
F
.
S
.
R
om
ol
o,
“
A
ddr
e
s
s
i
ng
f
or
e
ns
i
c
s
c
i
e
n
c
e
c
ha
l
l
e
nge
s
w
i
t
h
nuc
l
e
a
r
a
na
l
yt
i
c
a
l
t
e
c
hni
que
s
–
A
r
e
vi
e
w
,”
F
or
e
n
s
i
c
Sc
i
e
nc
e
I
nt
e
r
nat
i
onal
, vol
. 358, M
a
y 2024, d
oi
:
10.1016/
j
.f
or
s
c
i
i
nt
.2023.111767.
[
4]
Z
.
Z
ha
o
e
t
al
.
,
“
C
r
e
a
t
i
on
of
c
ot
t
on
m
ut
a
nt
l
i
br
a
r
y
ba
s
e
d
on
l
i
ne
a
r
e
l
e
c
t
r
on
a
c
c
e
l
e
r
a
t
or
r
a
di
a
t
i
on
m
ut
a
t
i
on,”
B
i
oc
he
m
i
s
t
r
y
an
d
B
i
ophy
s
i
c
s
R
e
por
t
s
, vol
. 30, J
ul
. 2022, doi
:
10.1016/
j
.bbr
e
p.2022.101228.
[
5]
L
. M
a
, F
. K
ong,
K
. S
un, T
. W
a
ng, a
nd T
. G
uo, “
F
r
om
c
l
a
s
s
i
c
a
l
r
a
di
a
t
i
on t
o m
ode
r
n r
a
di
a
t
i
on:
pa
s
t
, pr
e
s
e
nt
, a
nd f
ut
ur
e
of
r
a
di
a
t
i
on
m
ut
a
t
i
on br
e
e
di
ng,”
F
r
ont
i
e
r
s
i
n P
ubl
i
c
H
e
al
t
h
, vol
. 9, D
e
c
. 2021, doi
:
10.3389
/
f
pubh.2021.768071.
[
6]
J
.
Y
.
L
i
e
t
al
.
,
“
R
e
l
i
a
bi
l
i
t
y
a
na
l
ys
i
s
of
c
or
e
pow
e
r
opt
i
m
i
z
a
t
i
on
c
ont
r
ol
us
i
ng
K
a
l
m
a
n
f
i
l
t
e
r
f
o
r
a
c
c
e
l
e
r
a
t
or
dr
i
ve
n
s
ys
t
e
m
ba
s
e
d
on
r
e
c
onf
i
gur
a
bl
e
c
om
put
i
ng,”
A
nnal
s
of
N
uc
l
e
ar
E
ne
r
gy
, vol
. 192, N
ov. 2023, do
i
:
10.1016/
j
.a
nuc
e
ne
.2023.109959.
[
7]
X
.
Z
ha
ng,
N
.
P
u,
H
.
C
a
i
,
H
.
J
i
a
,
a
nd
Y
.
H
e
,
“
S
t
r
a
t
e
gi
e
s
of
e
l
i
m
i
na
t
i
ng
nuc
l
e
a
r
w
a
s
t
e
u
s
i
ng
a
c
c
e
l
e
r
a
t
or
-
dr
i
ve
n
s
ys
t
e
m
i
n
t
he
t
r
a
ns
i
t
i
on
s
t
a
ge
f
or
s
us
t
a
i
na
bl
e
a
nd
c
l
e
a
n
nuc
l
e
a
r
e
ne
r
gy
i
n
C
hi
na
,”
A
n
nal
s
of
N
uc
l
e
ar
E
ne
r
gy
,
vol
.
185,
J
un.
2023,
doi
:
10.1016/
j
.a
nuc
e
ne
.2023.109713.
[
8]
J
.
Y
.
L
i
e
t
al
.
,
“
T
he
opt
i
m
i
z
a
t
i
on
s
t
udy
o
f
c
or
e
pow
e
r
c
ont
r
ol
ba
s
e
d
on
m
e
t
a
-
he
ur
i
s
t
i
c
a
l
gor
i
t
hm
f
or
C
hi
na
i
n
i
t
i
a
t
i
ve
a
c
c
e
l
e
r
a
t
or
dr
i
ve
n
s
ubc
r
i
t
i
c
a
l
s
y
s
t
e
m
,”
N
uc
l
e
ar
E
ngi
ne
e
r
i
ng
and
T
e
c
hnol
ogy
,
vol
.
55,
no.
2,
pp.
452
–
459,
F
e
b.
2023
,
doi
:
10.1016/
j
.ne
t
.2022.10.004.
[
9]
J
. F
a
ng
e
t
al
.
, “
C
r
yoge
ni
c
s
e
c
onda
r
y e
l
e
c
t
r
on yi
e
l
d m
e
a
s
ur
e
m
e
nt
s
on
s
t
r
uc
t
ur
a
l
m
a
t
e
r
i
a
l
s
a
ppl
i
e
d i
n p
a
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
N
u
c
l
e
a
r
I
ns
t
r
um
e
nt
s
and
M
e
t
hods
i
n
P
hy
s
i
c
s
R
e
s
e
ar
c
h,
Se
c
t
i
on
A
:
A
c
c
e
l
e
r
at
or
s
,
Spe
c
t
r
om
e
t
e
r
s
,
D
e
t
e
c
t
o
r
s
and
A
s
s
oc
i
at
e
d
E
qui
pm
e
nt
,
vol
.
1027,
M
a
r
. 2022, doi
:
10.1016/
j
.ni
m
a
.2021.166292.
[
10]
E
.
T
r
a
c
ha
na
s
,
A
.
B
i
gna
m
i
,
N
.
G
a
z
i
s
,
K
.
M
i
c
he
l
,
a
nd
G
.
S
ol
br
e
kke
n,
“
E
va
l
ua
t
i
on
of
e
l
a
s
t
i
c
c
ons
t
a
nt
s
of
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
c
a
vi
t
y
m
a
t
e
r
i
a
l
s
w
i
t
h
r
e
s
on
a
nt
ul
t
r
a
s
ound
s
pe
c
t
r
os
c
opy,”
N
uc
l
e
ar
I
ns
t
r
um
e
nt
s
and
M
e
t
hods
i
n
P
hy
s
i
c
s
R
e
s
e
a
r
c
h,
Se
c
t
i
on
A
:
A
c
c
e
l
e
r
at
or
s
, Spe
c
t
r
om
e
t
e
r
s
, D
e
t
e
c
t
o
r
s
and A
s
s
o
c
i
at
e
d E
qui
pm
e
nt
, vol
. 1057, D
e
c
. 2023, doi
:
10.1016/
j
.ni
m
a
.2023.168791.
[
11]
L
.
R
os
s
i
,
“
P
a
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
a
nd
c
upr
a
t
e
s
upe
r
c
onduc
t
or
s
,”
P
hy
s
i
c
a
C
:
Supe
r
c
onduc
t
i
v
i
t
y
and
i
t
s
A
ppl
i
c
at
i
ons
,
vol
.
614,
N
ov. 2023, doi
:
10.1016/
j
.phys
c
.2023.1354360.
[
12]
D
.
M
or
ga
n,
G
.
P
i
l
a
ni
a
,
A
.
C
oue
t
,
B
.
P
.
U
be
r
ua
ga
,
C
.
S
un,
a
nd
J
.
L
i
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
i
n
nuc
l
e
a
r
m
a
t
e
r
i
a
l
s
r
e
s
e
a
r
c
h,”
C
ur
r
e
nt
O
pi
ni
on i
n Sol
i
d St
at
e
and M
at
e
r
i
al
s
Sc
i
e
nc
e
, vol
. 26, no. 2, A
pr
. 2022, doi
:
10.1016/
j
.c
os
s
m
s
.2021.100975.
[
13]
D
.
M
a
r
c
a
t
o,
D
.
B
or
t
ol
a
t
o,
V
.
M
a
r
t
i
ne
l
l
i
,
G
.
S
a
va
r
e
s
e
,
a
nd
G
.
A
.
S
us
t
o,
“
T
i
m
e
-
s
e
r
i
e
s
de
e
p
l
e
a
r
ni
ng
a
nom
a
l
y
de
t
e
c
t
i
on
f
or
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
I
F
A
C
-
P
ape
r
s
O
nL
i
ne
, vol
. 56, no. 2, pp. 1566
–
1571, 2023, doi
:
10.1016/
j
.i
f
a
c
ol
.2023.10.1855.
[
14]
C
.
A
l
l
a
i
r
e
e
t
al
.
,
“
A
r
t
i
f
i
c
i
a
l
I
nt
e
l
l
i
ge
nc
e
f
or
t
he
e
l
e
c
t
r
on
i
on
c
ol
l
i
de
r
(
A
I
4E
I
C
)
,”
C
om
put
i
ng
and
Sof
t
w
a
r
e
f
o
r
B
i
g
Sc
i
e
nc
e
,
vol
.
8,
no. 1, D
e
c
. 2024, doi
:
10.1007/
s
41781
-
024
-
00113
-
4.
[
15]
J
.
S
t
.
J
ohn
e
t
al
.
,
“
R
e
a
l
-
t
i
m
e
a
r
t
i
f
i
c
i
a
l
i
nt
e
l
l
i
ge
nc
e
f
or
a
c
c
e
l
e
r
a
t
or
c
ont
r
ol
:
a
s
t
udy
a
t
t
he
F
e
r
m
i
l
a
b
B
oos
t
e
r
,”
P
hy
s
i
c
al
R
e
v
i
e
w
A
c
c
e
l
e
r
at
or
s
and B
e
am
s
, vol
. 24, no. 10, O
c
t
. 2021, doi
:
10.1103/
P
hys
R
e
vA
c
c
e
l
B
e
a
m
s
.24.104601.
[
16]
T
.
D
or
i
go
e
t
al
.
,
“
T
ow
a
r
d
t
he
e
nd
-
to
-
e
nd
opt
i
m
i
z
a
t
i
on
of
pa
r
t
i
c
l
e
phys
i
c
s
i
ns
t
r
um
e
nt
s
w
i
t
h
di
f
f
e
r
e
nt
i
a
bl
e
pr
ogr
a
m
m
i
ng:
a
w
hi
t
e
pa
pe
r
,”
ar
X
i
v
-
P
hy
s
i
c
s
,
pp. 1
-
109, M
a
r
. 2022,
doi
:
10.48550/
a
r
X
i
v.2203.13818.
[
17]
R
.
Y
.
G
a
l
va
ni
,
N
.
E
f
f
e
ndy,
a
nd
A
.
K
us
um
a
w
a
nt
o,
“
E
va
l
ua
t
i
ng
w
e
i
ght
pr
i
or
i
t
y
on
gr
e
e
n
bui
l
di
ng
us
i
ng
f
uz
z
y
A
H
P
,”
2018
12t
h
Sout
h
E
as
t
A
s
i
an
T
e
c
hni
c
al
U
ni
v
e
r
s
i
t
y
C
on
s
or
t
i
um
(
SE
A
T
U
C
)
,
Y
ogya
ka
r
t
a
,
I
ndone
s
i
a
,
2018,
pp.
1
-
6,
doi
:
10.1109/
S
E
A
T
U
C
.2018.8788887
.
[
18]
S
.
N
a
f
i
s
a
h
a
nd
N
.
E
f
f
e
ndy,
“
V
oi
c
e
bi
om
e
t
r
i
c
s
ys
t
e
m
:
t
he
i
de
nt
i
f
i
c
a
t
i
on
of
t
h
e
s
e
ve
r
i
t
y
of
c
e
r
e
br
a
l
pa
l
s
y
us
i
ng
m
e
l
-
f
r
e
que
nc
i
e
s
s
t
oc
ha
s
t
i
c
s
a
ppr
oa
c
h
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
I
nt
e
gr
at
e
d
E
ngi
ne
e
r
i
ng
,
vol
.
11,
no.
3,
pp.
194
–
206,
S
e
p.
2019,
doi
:
10.30880/
i
j
i
e
.2019.11.03.020.
[
19]
D
.
E
.
P
.
L
e
buka
n,
A
.
N
.
I
.
W
a
r
da
na
,
a
nd
N
.
E
f
f
e
nd
y,
“
I
m
pl
e
m
e
nt
a
t
i
on
o
f
pl
a
nt
-
w
i
de
pi
-
f
uz
z
y
c
ont
r
ol
l
e
r
i
n
T
e
nne
s
s
e
e
E
a
s
t
m
a
n
pr
oc
e
s
s
,”
i
n
2019
I
n
t
e
r
nat
i
onal
Se
m
i
n
ar
on A
pp
l
i
c
a
t
i
on
f
or
T
e
c
hnol
ogy
of
I
n
f
or
m
at
i
on
and
C
o
m
m
uni
c
at
i
o
n
(
i
Se
m
an
t
i
c
)
,
S
e
m
a
r
a
n
g,
I
ndone
s
i
a
, 201
9, pp
. 450
-
45
4, do
i
:
1
0.110
9/
I
S
E
M
A
N
T
I
C
.2019
.8884
301
.
[
20]
D
.
H
.
K
i
m
e
t
al
.
,
“
D
e
e
p
n
e
ur
a
l
ne
t
w
or
k
-
ba
s
e
d
pr
e
di
c
t
i
on
f
or
l
ow
-
e
ne
r
gy
be
a
m
t
r
a
ns
por
t
t
uni
ng,”
J
ou
r
nal
of
t
he
K
o
r
e
an
P
hy
s
i
c
al
Soc
i
e
t
y
, vol
. 83, no. 8, pp. 647
–
653, O
c
t
. 2023, doi
:
10.1007/
s
40042
-
023
-
00848
-
0.
[
21]
S
.
H
.
G
u,
W
.
H
.
H
ua
ng,
C
.
Y
.
S
ong,
a
nd
W
.
T
.
Y
u,
“
M
ul
t
i
va
r
i
a
bl
e
vi
r
t
ua
l
di
a
gnos
t
i
c
s
a
nd
t
uni
ng
of
be
a
m
pos
i
t
i
oni
ng
us
i
n
g
m
a
c
hi
ne
l
e
a
r
ni
ng,”
N
uc
l
e
ar
I
ns
t
r
um
e
nt
s
and M
e
t
hods
i
n P
hy
s
i
c
s
R
e
s
e
ar
c
h, Se
c
t
i
on A
:
A
c
c
e
l
e
r
at
or
s
, Spe
c
t
r
om
e
t
e
r
s
,
D
e
t
e
c
t
or
s
an
d
A
s
s
oc
i
at
e
d E
qui
pm
e
nt
, vol
. 1052, J
ul
. 2023, doi
:
10.1016/
j
.ni
m
a
.2023.168245.
[
22]
A
.
S
c
he
i
nke
r
,
F
.
C
r
opp,
S
.
P
a
i
a
gua
,
a
nd
D
.
F
i
l
i
ppe
t
t
o,
“
A
n
a
da
pt
i
ve
a
ppr
oa
c
h
t
o
m
a
c
hi
ne
l
e
a
r
ni
ng
f
or
c
om
pa
c
t
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
, vol
. 11, no. 1, S
e
p. 2021, doi
:
10.1038/
s
41598
-
021
-
98785
-
0.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
14
, N
o.
4
,
A
ugus
t
20
25
:
3014
-
3021
3020
[
23]
P
.
A
r
pa
i
a
e
t
al
.
,
“
B
e
a
m
m
e
a
s
ur
e
m
e
nt
s
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng
a
t
t
he
c
e
r
n
l
a
r
ge
H
a
dr
on
C
ol
l
i
de
r
,”
I
E
E
E
I
ns
t
r
um
e
nt
at
i
on
and
M
e
as
ur
e
m
e
nt
M
agaz
i
ne
, vol
. 24, no. 9, pp. 47
–
58, D
e
c
. 2021, doi
:
10.1109/
M
I
M
.2021.9620023.
[
24]
N
.
E
f
f
e
ndy,
D
.
R
uhya
di
,
R
.
P
r
a
t
a
m
a
,
D
.
F
.
R
a
bba
,
A
.
F
.
A
ul
i
a
,
a
nd
A
.
Y
.
A
t
m
a
dj
a
,
“
F
or
e
s
t
qua
l
i
t
y
a
s
s
e
s
s
m
e
nt
ba
s
e
d
on
bi
r
d
s
ound
r
e
c
ogni
t
i
on
us
i
ng
c
onvol
ut
i
ona
l
ne
ur
a
l
ne
t
w
o
r
ks
,”
I
nt
e
r
nat
i
onal
J
our
nal
of
E
l
e
c
t
r
i
c
al
and
C
om
put
e
r
E
ngi
ne
e
r
i
ng
,
vol
.
12,
no. 4, pp. 4235
–
4242, A
ug. 2022, doi
:
10.11591/
i
j
e
c
e
.v12i
4.pp4235
-
4242.
[
25]
N
.
E
f
f
e
ndy,
N
.
C
.
W
a
c
hi
da
h,
B
.
A
c
hm
a
d,
P
.
J
i
w
a
ndono,
a
nd
M
.
S
ube
kt
i
,
“
P
ow
e
r
e
s
t
i
m
a
t
i
on
of
G
.A
.
S
i
w
a
b
e
s
s
y
m
ul
t
i
-
pur
pos
e
r
e
a
c
t
or
a
t
s
t
a
r
t
-
up
c
ondi
t
i
on
u
s
i
ng
a
r
t
i
f
i
c
i
a
l
ne
ur
a
l
n
e
t
w
or
k
w
i
t
h
i
nput
va
r
i
a
t
i
on,”
i
n
P
r
oc
e
e
di
ngs
-
2016
2nd
I
nt
e
r
nat
i
ona
l
C
onf
e
r
e
nc
e
on Sc
i
e
nc
e
and T
e
c
hnol
og
y
-
C
om
put
e
r
, I
C
ST
2016
, O
c
t
. 2017, pp. 1
33
–
138, doi
:
10.1109/
I
C
S
T
C
.2016.7877362.
[
26]
A
.
L
.
E
de
l
e
n,
S
.
G
.
B
i
e
dr
on,
B
. E
.
C
ha
s
e
,
D
.
E
ds
t
r
om
,
S
.
V
.
M
i
l
t
on,
a
nd
P
.
S
t
a
bi
l
e
,
“
N
e
ur
a
l
ne
t
w
or
ks
f
or
m
ode
l
i
ng
a
nd
c
ont
r
ol
o
f
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on
N
uc
l
e
ar
Sc
i
e
nc
e
,
vo
l
.
63,
no.
2,
pp.
878
–
897,
A
pr
.
2016
,
doi
:
10.1109/
T
N
S
.2016.2543203.
[
27]
P
.
R
e
c
h,
“
A
r
t
i
f
i
c
i
a
l
ne
ur
a
l
ne
t
w
or
ks
f
or
s
pa
c
e
a
nd
s
a
f
e
t
y
-
c
r
i
t
i
c
a
l
a
ppl
i
c
a
t
i
ons
:
r
e
l
i
a
bi
l
i
t
y
i
s
s
ue
s
a
nd
pot
e
nt
i
a
l
s
ol
ut
i
ons
,”
I
E
E
E
T
r
ans
ac
t
i
ons
on N
u
c
l
e
ar
Sc
i
e
nc
e
, vol
. 71, no. 4, pp. 377
–
404, A
pr
. 2024, doi
:
10.1109/
T
N
S
.2024.3349956.
[
28]
Y
.
M
.
R
e
n
e
t
al
.
,
“
A
t
ut
or
i
a
l
r
e
vi
e
w
of
ne
ur
a
l
ne
t
w
or
k
m
ode
l
i
ng
a
ppr
oa
c
h
e
s
f
or
m
ode
l
pr
e
di
c
t
i
ve
c
ont
r
ol
,”
C
om
put
e
r
s
and
C
he
m
i
c
al
E
ngi
ne
e
r
i
ng
, vol
. 165, S
e
p. 2022, doi
:
10.1016/
j
.c
om
pc
he
m
e
ng.2022
.107956.
[
29]
D
.
K
os
e
r
,
L
.
W
a
i
t
e
s
,
D
.
W
i
nkl
e
hn
e
r
,
M
.
F
r
e
y,
A
.
A
de
l
m
a
nn,
a
nd
J
.
C
onr
a
d,
“
I
nput
be
a
m
m
a
t
c
hi
ng
a
nd
be
a
m
dyna
m
i
c
s
de
s
i
g
n
opt
i
m
i
z
a
t
i
ons
of
t
he
I
s
oD
A
R
R
F
Q
us
i
ng
s
t
a
t
i
s
t
i
c
a
l
a
nd
m
a
c
hi
n
e
l
e
a
r
ni
ng
t
e
c
hni
que
s
,”
F
r
ont
i
e
r
s
i
n
P
hy
s
i
c
s
,
vol
.
10,
A
pr
.
2022,
doi
:
10.3389/
f
phy.2022.875889.
[
30]
A
.
A
s
l
a
m
e
t
a
l
.
,
“
N
e
u
r
a
l
ne
t
w
o
r
k
-
ba
s
e
d
c
on
t
r
o
l
o
f
a
n
u
l
t
r
a
f
a
s
t
l
a
s
e
r
,”
N
uc
l
e
ar
I
ns
t
r
u
m
e
n
t
s
an
d
M
e
t
h
o
ds
i
n
P
hy
s
i
c
s
R
e
s
e
ar
c
h
,
Se
c
t
i
o
n
A
:
A
c
c
e
l
e
r
a
t
o
r
s
,
S
pe
c
t
r
o
m
e
t
e
r
s
,
D
e
t
e
c
t
or
s
a
nd
A
s
s
oc
i
a
t
e
d
E
q
u
i
p
m
e
n
t
,
v
o
l
.
1
0
53
,
A
ug
.
2
02
3
,
do
i
:
1
0.
10
1
6/
j
.n
i
m
a
.2
02
3
.1
68
1
95
.
[
31]
K
.
S
un,
X
.
C
he
n,
X
.
Z
ha
o,
X
.
Q
i
,
Z
.
W
a
ng,
a
nd
Y
.
H
e
,
“
S
ur
r
oga
t
e
m
ode
l
of
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
us
i
ng
e
nc
ode
r
-
de
c
ode
r
ne
ur
a
l
ne
t
w
or
ks
w
i
t
h
phys
i
c
a
l
r
e
gul
a
r
i
z
a
t
i
on,”
I
nt
e
r
nat
i
onal
J
our
nal
of
M
ode
r
n
P
hy
s
i
c
s
A
,
vol
.
38,
no.
26
–
27,
S
e
p.
2023,
doi
:
10.1142/
S
0217751X
23501452.
[
32]
A
.
S
c
he
i
nke
r
,
F
.
C
r
opp,
a
nd
D
.
F
i
l
i
ppe
t
t
o,
“
A
da
pt
i
ve
a
ut
oe
nc
ode
r
l
a
t
e
nt
s
pa
c
e
t
uni
ng
f
o
r
m
or
e
r
obus
t
m
a
c
hi
ne
l
e
a
r
ni
ng
be
yond
t
he
t
r
a
i
ni
ng
s
e
t
f
or
s
i
x
-
di
m
e
ns
i
ona
l
pha
s
e
s
pa
c
e
di
a
gnos
t
i
c
s
of
a
t
i
m
e
-
va
r
yi
ng
ul
t
r
a
f
a
s
t
e
l
e
c
t
r
on
-
di
f
f
r
a
c
t
i
on
c
om
pa
c
t
a
c
c
e
l
e
r
a
t
or
,”
P
hy
s
i
c
al
R
e
v
i
e
w
E
, vol
. 107, no. 4, A
pr
. 2023, doi
:
10.1103/
P
hys
R
e
vE
.107.045302.
[
33]
F
.
M
.
V
e
l
ot
t
i
,
B
.
G
odda
r
d,
V
.
K
a
i
n,
R
.
R
a
m
j
i
a
w
a
n,
G
.
Z
.
D
.
P
or
t
a
,
a
nd
S
.
H
i
r
l
a
e
nde
r
,
“
T
ow
a
r
ds
a
ut
om
a
t
i
c
s
e
t
up
of
18
M
e
V
e
l
e
c
t
r
on
be
a
m
l
i
ne
us
i
ng
m
a
c
hi
ne
l
e
a
r
ni
ng,”
M
ac
hi
ne
L
e
ar
ni
ng:
Sc
i
e
nc
e
and
T
e
c
hnol
ogy
,
vol
.
4,
no.
2,
J
un.
2023,
doi
:
10.1088/
2632
-
2153/
a
c
c
e
21.
[
34]
X
.
C
he
n,
Y
.
J
i
a
,
X
.
Q
i
,
Z
.
W
a
ng,
a
nd
Y
.
H
e
,
“
O
r
bi
t
c
or
r
e
c
t
i
on
ba
s
e
d
on
i
m
p
r
ove
d
r
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
a
l
go
r
i
t
hm
,”
P
hy
s
i
c
al
R
e
v
i
e
w
A
c
c
e
l
e
r
at
or
s
and B
e
am
s
, vol
. 26, no. 4, A
pr
. 2023, doi
:
10.1103/
P
hys
R
e
vA
c
c
e
l
B
e
a
m
s
.26.044601.
[
35]
W
.
W
a
ng
e
t
al
.
,
“
A
c
c
e
l
e
r
a
t
e
d
de
e
p
r
e
i
nf
or
c
e
m
e
nt
l
e
a
r
ni
ng
f
or
f
a
s
t
f
e
e
dba
c
k
of
be
a
m
dyna
m
i
c
s
a
t
K
A
R
A
,
”
I
E
E
E
T
r
ans
ac
t
i
ons
on
N
uc
l
e
ar
Sc
i
e
n
c
e
, vol
. 68, no. 8, pp. 1794
–
1800, A
ug. 2021, doi
:
10.1109/
T
N
S
.2021.3084515.
[
36]
C
.
F
a
ne
l
l
i
e
t
al
.
,
“
A
I
-
a
s
s
i
s
t
e
d
opt
i
m
i
z
a
t
i
on
of
t
he
E
C
C
E
t
r
a
c
ki
ng
s
ys
t
e
m
a
t
t
h
e
E
l
e
c
t
r
on
I
on
C
ol
l
i
de
r
,”
N
uc
l
e
a
r
I
ns
t
r
um
e
nt
s
and
M
e
t
hods
i
n
P
hy
s
i
c
s
R
e
s
e
a
r
c
h,
Se
c
t
i
on
A
:
A
c
c
e
l
e
r
at
or
s
,
Spe
c
t
r
om
e
t
e
r
s
,
D
e
t
e
c
t
or
s
and
A
s
s
oc
i
at
e
d
E
qui
pm
e
nt
,
vol
.
1047,
F
e
b. 2023, doi
:
10.1016/
j
.ni
m
a
.2022.167748.
[
37]
J
.
O
.
L
übs
e
n,
M
.
S
c
hüt
t
e
,
S
.
S
c
hul
z
,
a
nd
A
.
E
i
c
hl
e
r
,
“
A
s
a
f
e
B
a
ye
s
i
a
n
o
pt
i
m
i
z
a
t
i
on
a
l
gor
i
t
hm
f
o
r
t
uni
ng
t
he
opt
i
c
a
l
s
ync
hr
oni
z
a
t
i
on
s
y
s
t
e
m
a
t
E
ur
ope
a
n
X
F
E
L
,”
I
F
A
C
-
P
ape
r
s
O
nL
i
ne
,
vol
.
56,
no.
2,
pp.
3079
–
3085,
2023
,
doi
:
10.1016/
j
.i
f
a
c
ol
.2023.10.1438.
[
38]
D
.
C
a
j
a
nde
r
,
D
.
A
gugl
i
a
,
I
.
V
i
a
r
ouge
,
a
nd
P
.
V
i
a
r
ouge
,
“
U
s
i
ng
s
upe
r
vi
s
e
d
m
a
c
hi
ne
l
e
a
r
ni
ng
i
n
pow
e
r
c
onve
r
t
e
r
s
de
s
i
gn
f
or
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
–
a
ppl
i
c
a
t
i
on
t
o
m
a
gne
t
i
c
c
om
pone
nt
s
de
s
i
gn,”
J
ou
r
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
,
vol
.
2687,
no.
8,
J
a
n. 2024, doi
:
10.1088/
1742
-
6596/
2687/
8/
082019.
[
39]
S
.
L
i
e
t
al
.
,
“
A
nove
l
a
ppr
oa
c
h
f
or
c
l
a
s
s
i
f
i
c
a
t
i
on
a
nd
f
or
e
c
a
s
t
i
ng
of
t
i
m
e
s
e
r
i
e
s
i
n
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
I
nf
or
m
at
i
on
,
vol
.
12
,
no. 3, M
a
r
. 2021, doi
:
10.3390/
i
nf
o12030121.
[
40]
E
.
K
r
ym
ova
,
G
.
O
boz
i
ns
ki
,
M
.
S
c
he
nk,
L
.
C
oyl
e
,
a
nd
T
.
P
i
e
l
oni
,
“
D
a
t
a
-
dr
i
ve
n
m
ode
l
i
ng
of
be
a
m
l
os
s
i
n
t
he
L
H
C
,”
F
r
ont
i
e
r
s
i
n
P
hy
s
i
c
s
, vol
. 10, J
a
n. 2023, doi
:
10.3389/
f
phy.2022.960963.
[
41]
C
.
E
m
m
a
,
A
.
E
de
l
e
n,
M
.
J
.
H
oga
n,
B
.
O
’
S
he
a
,
G
.
W
hi
t
e
,
a
nd
V
.
Y
a
ki
m
e
nko,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
-
ba
s
e
d
l
ongi
t
udi
na
l
pha
s
e
s
p
a
c
e
pr
e
di
c
t
i
on
of
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
s
,”
P
hy
s
i
c
al
R
e
v
i
e
w
A
c
c
e
l
e
r
at
o
r
s
and
B
e
am
s
,
vol
.
21,
no.
11,
N
ov.
2018,
doi
:
10.1103/
P
hys
R
e
vA
c
c
e
l
B
e
a
m
s
.21.112802.
[
42]
V
.
B
e
l
i
s
,
P
.
O
da
gi
u,
a
nd
T
.
K
.
A
a
r
r
e
s
t
a
d,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
f
or
a
nom
a
l
y
d
e
t
e
c
t
i
on
i
n
pa
r
t
i
c
l
e
phys
i
c
s
,”
R
e
v
i
e
w
s
i
n
P
hy
s
i
c
s
,
vol
.
12, D
e
c
. 2024, doi
:
10.1016/
j
.r
e
vi
p.2024.100091.
[
43]
Y
.
Y
u
e
t
al
.
,
“
I
ni
t
i
a
l
a
ppl
i
c
a
t
i
on
of
m
a
c
hi
ne
l
e
a
r
ni
ng
f
or
be
a
m
pa
r
a
m
e
t
e
r
opt
i
m
i
z
a
t
i
on
a
t
t
he
H
e
f
e
i
l
i
ght
s
o
ur
c
e
I
I
,”
J
ou
r
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
, vol
. 2687, no. 7,
J
a
n. 2024, doi
:
10.1088/
1742
-
6596/
2687/
7/
072002.
[
44]
Y
.
B
.
Y
u,
G
.
F
.
L
i
u,
W
.
X
u,
C
.
L
i
,
W
.
M
.
L
i
,
a
nd
K
.
X
ua
n,
“
R
e
s
e
a
r
c
h
on
t
u
ne
f
e
e
dba
c
k
of
t
he
H
e
f
e
i
l
i
ght
s
our
c
e
I
I
ba
s
e
d
on
m
a
c
hi
ne
l
e
a
r
ni
ng,”
N
uc
l
e
a
r
Sc
i
e
nc
e
and T
e
c
hni
qu
e
s
, vol
. 33, no. 3, M
a
r
. 2022, doi
:
10.1007/
s
41365
-
022
-
01018
-
w.
[
45]
T
.
J
e
s
ke
,
D
.
M
c
S
pa
dde
n,
N
.
K
a
l
r
a
,
T
.
B
r
i
t
t
on,
N
.
J
a
r
vi
s
,
a
nd
D
.
L
a
w
r
e
nc
e
,
“
A
I
f
o
r
e
xpe
r
i
m
e
nt
a
l
c
ont
r
ol
s
a
t
J
e
f
f
e
r
s
on
l
a
b,”
J
our
nal
of
I
ns
t
r
um
e
nt
at
i
on
, vol
. 17, no. 3, M
a
r
. 2022, doi
:
10.1088/
1748
-
0221/
17/
03/
C
03043.
[
46]
D
. S
c
hi
r
m
e
r
, “
A
m
a
c
hi
ne
l
e
a
r
ni
ng a
ppr
oa
c
h
t
o e
l
e
c
t
r
on or
bi
t
c
ont
r
ol
a
t
t
he
1.5
G
e
V
s
ync
hr
ot
r
on l
i
ght
s
our
c
e
D
E
L
T
A
,”
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
, vol
. 2420, no. 1, J
a
n. 2023, doi
:
10.1088/
1742
-
6596/
2420/
1/
012069.
[
47]
A
.
S
c
he
i
nke
r
,
“
A
da
pt
i
ve
c
ont
r
ol
a
nd
m
a
c
hi
ne
l
e
a
r
ni
ng
f
o
r
pa
r
t
i
c
l
e
a
c
c
e
l
e
r
a
t
or
be
a
m
c
ont
r
ol
a
nd
di
a
gnos
t
i
c
s
,”
P
r
oc
e
e
di
ngs
of
t
he
I
nt
e
r
nat
i
onal
B
e
am
I
ns
t
r
um
e
nt
at
i
on C
onf
e
r
e
nc
e
, I
B
I
C
, pp. 466
–
472, 2021, doi
:
10.18429/
J
A
C
oW
-
I
B
I
C
2021
-
T
H
O
B
03.
[
48]
T
.
G
a
l
l
a
ghe
r
,
A
.
W
ol
s
ki
,
a
nd
B
.
D
.
M
ur
a
t
o
r
i
,
“
A
m
a
c
hi
ne
l
e
a
r
ni
ng
a
ppr
oa
c
h
t
o
s
ha
pi
ng
m
a
gne
t
i
c
f
r
i
nge
f
i
e
l
ds
f
or
be
a
m
dyna
m
i
c
s
c
ont
r
ol
,”
J
our
nal
of
P
hy
s
i
c
s
:
C
onf
e
r
e
nc
e
Se
r
i
e
s
, vol
. 2687, no. 6, J
a
n. 2024, do
i
:
10.1088/
1742
-
6596/
2687/
6/
062031.
[
49]
A
.
H
a
nuka
e
t
al
.
,
“
A
c
c
ur
a
t
e
a
nd
c
onf
i
de
nt
pr
e
di
c
t
i
on
of
e
l
e
c
t
r
on
be
a
m
l
ongi
t
u
di
na
l
pr
ope
r
t
i
e
s
us
i
ng
s
pe
c
t
r
a
l
vi
r
t
ua
l
di
a
gnos
t
i
c
s
,”
Sc
i
e
nt
i
f
i
c
R
e
por
t
s
, vol
. 11, no. 1, F
e
b. 2021, doi
:
10.1038/
s
41598
-
021
-
82473
-
0.
[
50]
S
.
B
e
t
t
oni
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
ba
s
e
d
l
ongi
t
udi
na
l
vi
r
t
ua
l
di
a
gno
s
t
i
c
s
a
t
S
w
i
s
s
F
E
L
,”
R
e
v
i
e
w
of
Sc
i
e
nt
i
f
i
c
I
ns
t
r
um
e
nt
s
,
vol
. 95, no. 1, J
a
n. 2024, doi
:
10.1063/
5.0179712.
[
51]
R
.
L
i
,
Q
.
Z
ha
ng,
Z
.
Z
ha
o,
C
.
L
i
,
a
nd
K
.
W
a
ng,
“
R
e
s
e
a
r
c
h
on
be
a
m
dyna
m
i
c
s
opt
i
m
i
z
a
t
i
on
of
a
s
t
or
a
ge
r
i
ng
ba
s
e
d
on
m
a
c
hi
n
e
l
e
a
r
ni
ng,”
i
n
2023
I
nt
e
r
nat
i
onal
C
onf
e
r
e
nc
e
on
I
nt
e
l
l
i
ge
nt
C
om
put
i
ng
and
N
e
x
t
G
e
n
e
r
at
i
on
N
e
t
w
or
k
s
（
I
C
N
G
N
)
,
H
a
ng
z
hou,
C
hi
na
, 2023, pp. 1
-
5, doi
:
10.1109/
I
C
N
G
N
59831.2023.10396700
.
[
52]
P
.
A
r
pa
i
a
e
t
al
.
,
“
M
a
c
hi
ne
l
e
a
r
ni
ng
f
or
be
a
m
dyna
m
i
c
s
s
t
udi
e
s
a
t
t
he
C
E
R
N
L
a
r
ge
H
a
dr
on
C
ol
l
i
de
r
,”
N
uc
l
e
a
r
I
ns
t
r
u
m
e
nt
s
an
d
M
e
t
hods
i
n
P
hy
s
i
c
s
R
e
s
e
ar
c
h,
Se
c
t
i
on
A
:
A
c
c
e
l
e
r
at
o
r
s
,
Sp
e
c
t
r
om
e
t
e
r
s
,
D
e
t
e
c
t
or
s
and
A
s
s
o
c
i
at
e
d
E
qui
pm
e
nt
,
vol
.
985,
J
a
n. 2021, doi
:
10.1016/
j
.ni
m
a
.2020.164652.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
M
ac
hi
ne
l
e
ar
ni
ng appli
c
at
io
n f
or
par
ti
c
le
ac
c
e
le
r
at
or
opt
imi
z
at
io
n
-
a r
e
v
ie
w
(
I
s
ti
D
ia
n R
ac
hm
aw
at
i
)
3021
B
I
O
G
R
A
P
H
I
E
S
O
F
A
U
T
H
O
R
S
Isti
Dian
Rachmawati
received
a
B.Eng.
degree
in
Instrument
ation
Electronics
from
the
Polytechnic
Institute
of
Nuclear
Technology
in
2012.
She
is
pursuing
a
master'
s
in
the
Department
of
Nuclear
Engineering
and
Engineering
Physics,
Faculty
of
Engineering,
Universitas
Gadjah
Mada.
She
is
a
research
assistant
at
the
Intellige
nt
and
Embedded
System
Resear
ch
Group,
Depar
tment
of
Nuclea
r
Enginee
ring
and
Engine
ering
Physics,
Faculty
of
Engineering,
Universitas
Gadjah
Mada,
and
a
junior
researcher
at
the
Research
Center
for
Accelerator
Technology,
National
Research,
and
Innovation
Agency
o
f
Indonesia.
Her
research
interests
are
machine
learning
and
its
application
in
engineer
ing.
She
can
be
contacte
d
at
email:
isti004@
brin.go.id.
Nazrul
Effendy
received
a
B.Eng.
degree
in
Instrumentat
ion
Technology
o
f
Nuclear
Engineering
and
an
M.Eng.
degree
in
Electrical
Engineering
from
Universitas
Gadjah
Mada
in
1998
and
2001.
He
received
a
Ph.D.
degree
in
Ele
ctrical
Engineering
from
Chulalongk
orn
Universi
ty
in
2009.
He
was
a
resea
rch
fellow
at
the
Depar
tment
of
Control
and
Computer
Enginee
ring,
the
Polytechni
c
Universi
ty
of
Turin,
in
201
0
and
2011
and
a
visiting
researcher
in
the
Shinoda
Lab
(
pattern
re
cognition
&
its
applicatio
ns
to
real
world
),
Tokyo
Institut
e
of
Technology
in
2009.
He
is
a
professor
and
the
coordin
ator
of
the
Intelligent
and
Embedded
System
Research
Group
in
the
Department
of
Nuclear
En
gineering
and
Engineering
Physics,
Faculty
of
Engine
ering,
Univer
sitas
Gadja
h
Mada.
He
is
a
member
of
the
Indone
sian
Association
of
Pattern
Recognition,
the
Indonesian
Society
for
Soft
Computing,
the
Indonesian
Artificial
Intelligence
Society,
and
the
International
As
sociation
for
P
attern
Recognition.
He
can
be contacted at email: nazr
ul@
ugm.ac.
id.
Taufik
received
a
B.Eng.
degree
in
Physics
from
Universitas
Pad
jadjaran
in
2004
and
an
M.Sc.
degree
in
Physics
from
Universitas
Gadjah
Mada
in
2
013.
He
received
a
Ph.D.
degree
in
Accelerator
Science
from
Sokendai
Tsukuba,
Japan,
in
201
9.
He
is
a
senior
researche
r
at
the
Resear
ch
Center
for
Accel
erator
Technol
ogy,
Resear
ch
Organi
zation
for
Nuclea
r
Energy,
National
Research
and
Innovation
Agency
of
Indonesia
.
He
is
the
coordinator
of
the
linear
accelerator
research
group
.
His
research
interests
are
accelerator
scie
nce
and
machine
learning.
He can be contacted at email:
tauf009@
brin.go.id
.
Evaluation Warning : The document was created with Spire.PDF for Python.