Mathematics,
Год журнала:
2025,
Номер
13(4), С. 675 - 675
Опубликована: Фев. 18, 2025
With
the
rapid
advancement
of
artificial
intelligence
(AI)
technology,
demand
for
vast
amounts
data
training
AI
algorithms
to
attain
has
become
indispensable.
However,
in
realm
big
high
feature
dimensions
frequently
give
rise
overfitting
issues
during
training,
thereby
diminishing
model
accuracy.
To
enhance
prediction
accuracy,
selection
(FS)
methods
have
arisen
with
goal
eliminating
redundant
features
within
datasets.
In
this
paper,
a
highly
efficient
FS
method
advanced
performance,
called
EMEPO,
is
proposed.
It
combines
three
learning
strategies
on
basis
Parrot
Optimizer
(PO)
better
ensure
performance.
Firstly,
novel
exploitation
strategy
introduced,
which
integrates
randomness,
optimality,
and
Levy
flight
algorithm’s
local
capabilities,
reduce
execution
time
solving
problems,
classification
Secondly,
multi-population
evolutionary
takes
into
account
diversity
individuals
based
fitness
values
optimize
balance
between
exploration
stages
algorithm,
ultimately
improving
capability
explore
solution
space
globally.
Finally,
unique
focusing
individual
boost
population
problems.
This
approach
improves
capacity
avoid
suboptimal
subsets.
The
EMEPO-based
tested
23
datasets
spanning
low-,
medium-,
high-dimensional
data.
results
show
exceptional
performance
reduction,
efficiency,
convergence
speed,
stability.
indicates
promise
as
an
effective
selection.
Artificial Intelligence Review,
Год журнала:
2024,
Номер
57(5)
Опубликована: Апрель 23, 2024
Abstract
This
study
introduces
a
novel
population-based
metaheuristic
algorithm
called
secretary
bird
optimization
(SBOA),
inspired
by
the
survival
behavior
of
birds
in
their
natural
environment.
Survival
for
involves
continuous
hunting
prey
and
evading
pursuit
from
predators.
information
is
crucial
proposing
new
that
utilizes
abilities
to
address
real-world
problems.
The
algorithm's
exploration
phase
simulates
snakes,
while
exploitation
models
escape
During
this
phase,
observe
environment
choose
most
suitable
way
reach
secure
refuge.
These
two
phases
are
iteratively
repeated,
subject
termination
criteria,
find
optimal
solution
problem.
To
validate
performance
SBOA,
experiments
were
conducted
assess
convergence
speed,
behavior,
other
relevant
aspects.
Furthermore,
we
compared
SBOA
with
15
advanced
algorithms
using
CEC-2017
CEC-2022
benchmark
suites.
All
test
results
consistently
demonstrated
outstanding
terms
quality,
stability.
Lastly,
was
employed
tackle
12
constrained
engineering
design
problems
perform
three-dimensional
path
planning
Unmanned
Aerial
Vehicles.
demonstrate
that,
contrasted
optimizers,
proposed
can
better
solutions
at
faster
pace,
showcasing
its
significant
potential
addressing
International Journal of Systems Science,
Год журнала:
2024,
Номер
55(15), С. 3185 - 3222
Опубликована: Июль 1, 2024
In
recent
research,
metaheuristic
strategies
stand
out
as
powerful
tools
for
complex
optimization,
capturing
widespread
attention.
This
study
proposes
the
Educational
Competition
Optimizer
(ECO),
an
algorithm
created
diverse
optimization
tasks.
ECO
draws
inspiration
from
competitive
dynamics
observed
in
real-world
educational
resource
allocation
scenarios,
harnessing
this
principle
to
refine
its
search
process.
To
further
boost
efficiency,
divides
iterative
process
into
three
distinct
phases:
elementary,
middle,
and
high
school.
Through
stepwise
approach,
gradually
narrows
down
pool
of
potential
solutions,
mirroring
gradual
competition
witnessed
within
systems.
strategic
approach
ensures
a
smooth
resourceful
transition
between
ECO's
exploration
exploitation
phases.
The
results
indicate
that
attains
peak
performance
when
configured
with
population
size
40.
Notably,
algorithm's
efficacy
does
not
exhibit
strictly
linear
correlation
size.
comprehensively
evaluate
effectiveness
convergence
characteristics,
we
conducted
rigorous
comparative
analysis,
comparing
against
nine
state-of-the-art
algorithms.
remarkable
success
efficiently
addressing
problems
underscores
applicability
across
domains.
additional
resources
open-source
code
proposed
can
be
accessed
at
https://aliasgharheidari.com/ECO.html
https://github.com/junbolian/ECO.
Artificial Intelligence Review,
Год журнала:
2025,
Номер
58(3)
Опубликована: Янв. 6, 2025
The
advent
of
the
intelligent
information
era
has
witnessed
a
proliferation
complex
optimization
problems
across
various
disciplines.
Although
existing
meta-heuristic
algorithms
have
demonstrated
efficacy
in
many
scenarios,
they
still
struggle
with
certain
challenges
such
as
premature
convergence,
insufficient
exploration,
and
lack
robustness
high-dimensional,
nonconvex
search
spaces.
These
limitations
underscore
need
for
novel
techniques
that
can
better
balance
exploration
exploitation
while
maintaining
computational
efficiency.
In
response
to
this
need,
we
propose
Artificial
Lemming
Algorithm
(ALA),
bio-inspired
metaheuristic
mathematically
models
four
distinct
behaviors
lemmings
nature:
long-distance
migration,
digging
holes,
foraging,
evading
predators.
Specifically,
migration
burrow
are
dedicated
highly
exploring
domain,
whereas
foraging
predators
provide
during
process.
addition,
ALA
incorporates
an
energy-decreasing
mechanism
enables
dynamic
adjustments
between
exploitation,
thereby
enhancing
its
ability
evade
local
optima
converge
global
solutions
more
robustly.
To
thoroughly
verify
effectiveness
proposed
method,
is
compared
17
other
state-of-the-art
on
IEEE
CEC2017
benchmark
test
suite
CEC2022
suite.
experimental
results
indicate
reliable
comprehensive
performance
achieve
superior
solution
accuracy,
convergence
speed,
stability
most
cases.
For
29
10-,
30-,
50-,
100-dimensional
functions,
obtains
lowest
Friedman
average
ranking
values
among
all
competitor
methods,
which
1.7241,
2.1034,
2.7241,
2.9310,
respectively,
12
again
wins
optimal
2.1667.
Finally,
further
evaluate
applicability,
implemented
address
series
cases,
including
constrained
engineering
design,
photovoltaic
(PV)
model
parameter
identification,
fractional-order
proportional-differential-integral
(FOPID)
controller
gain
tuning.
Our
findings
highlight
competitive
edge
potential
real-world
applications.
source
code
publicly
available
at
https://github.com/StevenShaw98/Artificial-Lemming-Algorithm
.
Artificial Intelligence Review,
Год журнала:
2025,
Номер
58(3)
Опубликована: Янв. 6, 2025
Abstract
Optimization
algorithms
play
a
crucial
role
in
solving
complex
challenges
across
various
fields,
including
engineering,
finance,
and
data
science.
This
study
introduces
novel
hybrid
optimization
algorithm,
the
Hybrid
Crayfish
Algorithm
with
Differential
Evolution
(HCOADE),
which
addresses
limitations
of
premature
convergence
inadequate
exploitation
traditional
(COA).
By
integrating
COA
(DE)
strategies,
HCOADE
leverages
DE’s
mutation
crossover
mechanisms
to
enhance
global
performance.
The
COA,
inspired
by
foraging
social
behaviors
crayfish,
provides
flexible
framework
for
exploring
solution
space,
while
robust
strategies
effectively
exploit
this
space.
To
evaluate
HCOADE’s
performance,
extensive
experiments
are
conducted
using
34
benchmark
functions
from
CEC
2014
2017,
as
well
six
engineering
design
problems.
results
compared
ten
leading
algorithms,
classical
Particle
Swarm
(PSO),
Grey
Wolf
Optimizer
(GWO),
Whale
(WOA),
Moth-flame
(MFO),
Salp
(SSA),
Reptile
Search
(RSA),
Sine
Cosine
(SCA),
Constriction
Coefficient-Based
Gravitational
(CPSOGSA),
Biogeography-based
(BBO).
average
rankings
Wilcoxon
Rank
Sum
Test
provide
comprehensive
comparison
clearly
demonstrating
its
superiority.
Furthermore,
performance
is
assessed
on
2020
2022
test
suites,
further
confirming
effectiveness.
A
comparative
analysis
against
notable
winners
competitions,
LSHADEcnEpSin,
LSHADESPACMA,
CMA-ES,
CEC-2017
suite,
revealed
superior
HCOADE.
underscores
advantages
DE
offers
valuable
insights
addressing
Biomimetics,
Год журнала:
2025,
Номер
10(1), С. 23 - 23
Опубликована: Янв. 3, 2025
To
address
the
challenges
of
slow
convergence
speed,
poor
precision,
and
getting
stuck
in
local
optima
for
unmanned
aerial
vehicle
(UAV)
three-dimensional
path
planning,
this
paper
proposes
a
planning
method
based
on
an
Improved
Human
Evolution
Optimization
Algorithm
(IHEOA).
First,
mathematical
model
is
used
to
construct
terrain
environment,
multi-constraint
cost
established,
framing
as
multidimensional
function
optimization
problem.
Second,
recognizing
sensitivity
population
diversity
Logistic
Chaotic
Mapping
traditional
(HEOA),
opposition-based
learning
strategy
employed
uniformly
initialize
distribution,
thereby
enhancing
algorithm’s
global
capability.
Additionally,
guidance
factor
introduced
into
leader
role
during
development
stage,
providing
clear
directionality
search
process,
which
increases
probability
selecting
optimal
paths
accelerates
speed.
Furthermore,
loser
update
strategy,
adaptive
t-distribution
perturbation
utilized
its
small
mutation
amplitude,
enhances
capability
robustness
algorithm.
Evaluations
using
12
standard
test
functions
demonstrate
that
these
improvement
strategies
effectively
enhance
precision
algorithm
stability,
with
IHEOA,
integrates
multiple
strategies,
performing
particularly
well.
Experimental
comparative
research
three
different
environments
five
algorithms
shows
IHEOA
not
only
exhibits
excellent
performance
terms
speed
but
also
generates
superior
while
demonstrating
exceptional
complex
environments.
These
results
validate
significant
advantages
proposed
improved
addressing
UAV
challenges.
