liacs natural computing group leiden university evolutionary algorithms
TRANSCRIPT
![Page 1: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/1.jpg)
LIACS Natural Computing Group Leiden University
Evolutionary Algorithms
![Page 2: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/2.jpg)
LIACS Natural Computing Group Leiden University2
Overview
• Introduction: Optimization and EAs• Genetic Algorithms• Evolution Strategies• Theory• Examples
![Page 3: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/3.jpg)
LIACS Natural Computing Group Leiden University3
Background IBiology = Engineering (Daniel Dennett)
![Page 4: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/4.jpg)
LIACS Natural Computing Group Leiden University4
Introduction • Modeling
• Simulation
• Optimization
! ! ! ! ! !???
! ! ! ???! ! !
??? ! ! !! ! !
Input: Known (measured)Output: Known (measured)
Interrelation: Unknown
Input: Will be given
Model: Already exists
How is the result for the input?
Objective: Will be givenHow (with which parameter settings)to achieve this objective?
![Page 5: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/5.jpg)
LIACS Natural Computing Group Leiden University5
Simulation vs. Optimization
SimulatorSimulator
… what happens if?
Result
Trial & Error
SimulatorSimulator
... how do I achieve the best result?
OptimalResult
Maximization / MinimizationIf so, multiple objectives
OptimizerOptimizer
![Page 6: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/6.jpg)
LIACS Natural Computing Group Leiden University6
Introduction:
OptimizationEvolutionary Algorithms
![Page 7: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/7.jpg)
LIACS Natural Computing Group Leiden University7
Optimization
• f : objective function
– High-dimensional
– Non-linear, multimodal
– Discontinuous, noisy, dynamic
• M M1 M2 ... Mn heterogeneous
• Restrictions possible over
• Good local, robust optimum desired
• Realistic landscapes are like that!Global Minimum
Local, robust optimum
![Page 8: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/8.jpg)
LIACS Natural Computing Group Leiden University8
• Illustrative Example: Optimize Efficiency– Initial:
– Evolution:
• 32% Improvement in Efficiency !
Optimization Creating Innovation
![Page 9: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/9.jpg)
LIACS Natural Computing Group Leiden University9
Dynamic Optimization
• Dynamic Function• 30-dimensional• 3D-Projection
![Page 10: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/10.jpg)
LIACS Natural Computing Group Leiden University10
Classification of Optimization Algorithms• Direct optimization algorithm:
Evolutionary Algorithms
• First order optimization algorithm:
e.g., gradient method
• Second order optimization algorithm:
e.g., Newton method
![Page 11: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/11.jpg)
LIACS Natural Computing Group Leiden University11
Iterative Optimization Methods
General description:Actual Point
New Point
Directional vector
Step size (scalar)
1tx
tx
tt vs
1x
2x
3x• At every Iteration:
– Choose direction
– Determine step size
• Direction:
– Gradient
– Random
• Step size:
– 1-dim. optimization
– Random
– Self-adaptive
![Page 12: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/12.jpg)
LIACS Natural Computing Group Leiden University12
• Global convergence with probability one:
• General, but for practical purposes useless
• Convergence velocity:
• Local analysis only, specific (convex) functions
1))(*Pr(lim
tPxt
)))(())1((( maxmax tPftPfE
The Fundamental Challenge
![Page 13: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/13.jpg)
LIACS Natural Computing Group Leiden University13
• Global convergence (with probability 1):
• General statement (holds for all functions)
• Useless for practical situations:
– Time plays a major role in practice
– Not all objective functions are relevant in practice
1))(*Pr(lim
tPxt
Theoretical Statements
![Page 14: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/14.jpg)
LIACS Natural Computing Group Leiden University14
x1
x2
f(x1,x2)
(x*1,x*2)
f(x*1,x*2)
An Infinite Number of Pathological Cases !
• NFL-Theorem:
– All optimization algorithms perform equally well iff performance is averaged over all possible optimization problems.
• Fortunately: We are not Interested in „all possible problems“
![Page 15: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/15.jpg)
LIACS Natural Computing Group Leiden University15
• Convergence velocity:
• Very specific statements
– Convex objective functions
– Describes convergence in local optima
– Very extensive analysis for Evolution Strategies
)))(())1((( maxmax tPftPfE
Theoretical Statements
![Page 16: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/16.jpg)
LIACS Natural Computing Group Leiden University
Evolutionary AlgorithmPrinciples
![Page 17: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/17.jpg)
LIACS Natural Computing Group Leiden University17
Model-Optimization-Action
Evaluation
Optimizer
Business Process Model
Simulation
215
1
i
iii
i scale
desiredcalculatedweightquality
Function Model from Data
Experiment SubjectiveFunction(s)
...)( yfi
![Page 18: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/18.jpg)
LIACS Natural Computing Group Leiden University18
Evolutionary Algorithms TaxonomyEvolutionary AlgorithmsEvolutionary Algorithms
Evolution StrategiesEvolution Strategies Genetic AlgorithmsGenetic Algorithms OtherOther
• Evolutionary Progr.• Differential Evol.• GP• PSO• EDA• Real-coded Gas• …
• Mixed-integer capabilities
• Emphasis on mutation
• Self-adaptation
• Small population sizes
• Deterministic selection
• Developed in Germany
• Theory focused on convergence velocity
• Discrete representations
• Emphasis on crossover
• Constant parameters
• Larger population sizes
• Probabilistic selection
• Developed in USA
• Theory focused on schema processing
![Page 19: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/19.jpg)
LIACS Natural Computing Group Leiden University19
Generalized Evolutionary Algorithm
t := 0;
initialize(P(t));
evaluate(P(t));
while not terminate do
P‘(t) := mating_selection(P(t));
P‘‘(t) := variation(P‘(t));
evaluate(P‘‘(t));
P(t+1) := environmental_selection(P‘‘(t) Q);
t := t+1;
od
![Page 20: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/20.jpg)
LIACS Natural Computing Group Leiden University
Genetic Algorithms
![Page 21: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/21.jpg)
LIACS Natural Computing Group Leiden University21
Genetic Algorithms: Mutation
• Mutation by bit inversion with probability pm.
• pm identical for all bits.
• pm small (e.g., pm = 1/n).
0 1 1 1 0 1 0 1 0 0 0 0 0 01
0 1 1 1 0 0 0 1 0 1 0 0 0 01
![Page 22: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/22.jpg)
LIACS Natural Computing Group Leiden University22
Genetic Algorithms: Crossover
• Crossover applied with probability pc.
• pc identical for all individuals.
• k-point crossover: k points chosen randomly.• Example: 2-point crossover.
![Page 23: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/23.jpg)
LIACS Natural Computing Group Leiden University23
Genetic Algorithms: Selection
• Fitness proportional:– f fitness, population size
• Tournament selection:– Randomly select q << individuals.– Copy best of these q into next generation.– Repeat times.– q is the tournament size (often: q = 2).
![Page 24: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/24.jpg)
LIACS Natural Computing Group Leiden University
Some Theory
![Page 25: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/25.jpg)
LIACS Natural Computing Group Leiden University25
l
iiaaf
1
)(
jfljfl
ij
aifif
i
aa
a
a
k
k
a
a
a
a
qpj
flqp
i
fkp
kafamfkp
kpk
10
0)11(
)(
))())((Pr()(
)(max
Convergence Velocity Analysis• (1+1)-GA, (1,)-GA, (1+)-GA• For counting ones function• Convergence velocity:
![Page 26: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/26.jpg)
LIACS Natural Computing Group Leiden University26
llafp
1
)1)((2
1*
QR
IP
0
Tj
iji ntE )(
1)()( QInN ij
• Optimum mutation rate ?
• Absorption times from transition matrix
in block form, using
Convergence Velocity Analysis II
![Page 27: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/27.jpg)
LIACS Natural Computing Group Leiden University27
p
Convergence Velocity Analysis III• P too large:
– Exponential• P too small:
– Almost constant• Optimal:
– O(l ln l)
![Page 28: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/28.jpg)
LIACS Natural Computing Group Leiden University28
ikk
ikk
i
fl
kk
ppi
ka
''
1)1(
min,
• (1,l)-GA (kmin = -fa), (1+l)-GA (kmin = 0) :
Convergence Velocity Analysis IV
![Page 29: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/29.jpg)
LIACS Natural Computing Group Leiden University29
Convergence Velocity Analysis V• (1,)-GA, (1GA (1,)-ES, (1ES
• Conclusion: Unifying, search-space independent theory ?
![Page 30: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/30.jpg)
LIACS Natural Computing Group Leiden University30
)(1
1)(
min,
kpkafl
kk
jkk
ikk
jikk
ji
ppp
j
i
ikp
'1
'1
'
0
1
1
)1(
1)(
Convergence Velocity Analysis VI• (,)-GA, (GA
• Theory• Experiment
![Page 31: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/31.jpg)
LIACS Natural Computing Group Leiden University
Evolution Strategies
![Page 32: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/32.jpg)
LIACS Natural Computing Group Leiden University32
Evolution Strategy – Basics
• Mostly real-valued search space IRn
– also mixed-integer, discrete spaces• Emphasis on mutation
– n-dimensional normal distribution– expectation zero
• Different recombination operators• Deterministic selection
– ()-selection: Deterioration possible– ()-selection: Only accepts improvements
• , i.e.: Creation of offspring surplus• Self-adaptation of strategy parameters.
![Page 33: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/33.jpg)
LIACS Natural Computing Group Leiden University33
Representation of search points
)),,...,(( 1 nxxa
),...,( 1 nxxa
• Simple ES with 1/5 success rule:
– Exogenous adaptation of step size
– Mutation: N(0, )
• Self-adaptive ES with single step size:
– One controls mutation for all xi
– Mutation: N(0, )
![Page 34: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/34.jpg)
LIACS Natural Computing Group Leiden University34
Representation of search points
)),...,(),,...,(( 11 nnxxa
)),...,(),,...,(),,...,(( 2/)1(111 nnnnxxa
• Self-adaptive ES with individual step sizes:
– One individual i per xi
– Mutation: Ni(0, i)
• Self-adaptive ES with correlated mutation:
– Individual step sizes
– One correlation angle per coordinate pair
– Mutation according to covariance matrix: N(0, C)
![Page 35: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/35.jpg)
LIACS Natural Computing Group Leiden University35
Evolution Strategy:
AlgorithmsMutation
![Page 36: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/36.jpg)
LIACS Natural Computing Group Leiden University36
Operators: Mutation – one
)1,0(
))1,0(exp(
)),,...,((
)),,...,((
0
1
1
iii
n
n
Nxx
N
xxa
xxa
• Self-adaptive ES with one step size:
– One controls mutation for all xi
– Mutation: N(0, )Individual before mutation
Individual after mutation
1.: Mutation of step sizes
2.: Mutation of objective variables
Here the new ‘ is used!
![Page 37: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/37.jpg)
LIACS Natural Computing Group Leiden University37
Operators: Mutation – one
n
10
• Thereby 0 is the so-called learning rate
– Affects the speed of the -Adaptation
– 0 bigger: faster but more imprecise
– 0 smaller: slower but more precise
– How to choose 0?
– According to recommendation of Schwefel*:
*H.-P. Schwefel: Evolution and Optimum Seeking, Wiley, NY, 1995.
![Page 38: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/38.jpg)
LIACS Natural Computing Group Leiden University38
Operators: Mutation – one
Position of parents (here: 5)
Offspring of parent lies on the hyper sphere (for n > 10);Position is uniformly distributed
Contour lines of objective function
![Page 39: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/39.jpg)
LIACS Natural Computing Group Leiden University39
Pros and Cons: One • Advantages:
– Simple adaptation mechanism
– Self-adaptation usually fast and precise
• Disadvantages:
– Bad adaptation in case of complicated contour lines
– Bad adaptation in case of very differently scaled object variables
• -100 < xi < 100 and e.g. -1 < xj < 1
![Page 40: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/40.jpg)
LIACS Natural Computing Group Leiden University40
Operators: Mutation – individual i
)1,0(
))1,0()1,0(exp(
)),...,(),,...,((
)),...,(),,...,((
11
11
iiii
iii
nn
nn
Nxx
NN
xxa
xxa
• Self-adaptive ES with individual step sizes:
– One i per xi
– Mutation: Ni(0, i) Individual before Mutation
Individual after Mutation
1.: Mutation of individual step sizes
2.: Mutation of object variables
The new individual i‘ are used here!
![Page 41: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/41.jpg)
LIACS Natural Computing Group Leiden University41
Operators: Mutation – individual i
nn 2
1
2
1
• , ‘ are learning rates, again– ‘: Global learning rate– N(0,1): Only one realisation– : local learning rate
– Ni(0,1): n realisations
– Suggested by Schwefel*:
*H.-P. Schwefel: Evolution and Optimum Seeking, Wiley, NY, 1995.
![Page 42: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/42.jpg)
LIACS Natural Computing Group Leiden University42
Position of parents (here: 5)
Offspring are located on the hyperellipsoid (für n > 10);position equally distributed.
Contour lines
Operators: Mutation – individual i
![Page 43: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/43.jpg)
LIACS Natural Computing Group Leiden University43
Pros and Cons: Individual i
• Advantages:
– Individual scaling of object variables
– Increased global convergence reliability
• Disadvantages:
– Slower convergence due to increased learning effort
– No rotation of coordinate system possible• Required for badly conditioned objective function
![Page 44: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/44.jpg)
LIACS Natural Computing Group Leiden University44
Operators: Correlated Mutations
),0(
)1,0(
))1,0()1,0(exp(
)),...,(),,...,(),,...,((
)),...,(),,...,(),,...,((
2/)1(111
2/)1(111
CNxx
N
NN
xxa
xxa
ii
jjj
iii
nnnn
nnnn
Individual before mutation
Individual after mutation
1.: Mutation of Individual step sizes
2.: Mutation of rotation angles
New convariance matrix C‘ used here!
• Self-adaptive ES with correlated mutations:
– Individual step sizes
– One rotation angle for each pair of coordinates
– Mutation according to covariance matrix: N(0, C)
3.: Mutation of object variables
![Page 45: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/45.jpg)
LIACS Natural Computing Group Leiden University45
Operators: Correlated Mutations
)2tan()(2
1 22)( ijjijiijc
• Interpretation of rotation angles ij
• Mapping onto convariances according to
x1
x2
1
2 12
![Page 46: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/46.jpg)
LIACS Natural Computing Group Leiden University46
Operators: Correlated Mutation
)(2 jjjj sign
, ‘, are again learning rates ‘ as before = 0,0873 (corresponds to 5 degree) Out of boundary correction:
![Page 47: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/47.jpg)
LIACS Natural Computing Group Leiden University47
Correlated Mutations for ES
Position of parents (hier: 5)
Offspring is located on the Rotatable hyperellipsoid(for n > 10); position equallydistributed.
Contour lines
![Page 48: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/48.jpg)
LIACS Natural Computing Group Leiden University48
Operators: Correlated Mutations),0( CN
• How to create ?
– Multiplication of uncorrelated mutation vector with n(n-1)/2 rotational matrices
– Generates only feasible (positiv definite) correlation matrices
u
n
i
n
ijijc R
)(1
1 1
![Page 49: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/49.jpg)
LIACS Natural Computing Group Leiden University49
Operators: Correlated Mutations• Structure of rotation matrix
1
1
)cos(
1
)sin(
)sin(
1
)cos(1
1
)(
0
0
ijij
ijij
ijR
![Page 50: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/50.jpg)
LIACS Natural Computing Group Leiden University50
Operators: Correlated Mutations• Implementation of correlated mutations
nq := n(n-1)/2;for i:=1 to n do
u[i] := [i] * Ni(0,1);
for k:=1 to n-1 don1 := n-k;n2 := n;for i:=1 to k do
d1 := u[n1]; d2:= u[n2];u[n2] := d1*sin([nq])+ d2*cos([nq]);u[n1] := d1*cos([nq])- d2*sin([nq]);n2 := n2-1;nq := nq-1;
odod
Generation of the uncorrelated mutation vector
Rotations
![Page 51: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/51.jpg)
LIACS Natural Computing Group Leiden University51
Pros and Cons: Correlated Mutations
• Advantages:
– Individual scaling of object variables
– Rotation of coordinate system possible
– Increased global convergence reliability
• Disadvantages:
– Much slower convergence
– Effort for mutations scales quadratically
– Self-adaptation very inefficient
![Page 52: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/52.jpg)
LIACS Natural Computing Group Leiden University52
Operators: Mutation – Addendum
• Generating N(0,1)-distributed random numbers?
w
wvx
w
wux
vuw
Uv
Uu
)log(2
)log(2
1)1,0[2
1)1,0[2
2
1
22
If w > 1
)1,0(~, 21 Nxx
![Page 53: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/53.jpg)
LIACS Natural Computing Group Leiden University53
Evolution Strategy:
AlgorithmsRecombination
![Page 54: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/54.jpg)
LIACS Natural Computing Group Leiden University54
Operators: Recombination• Only for > 1
• Directly after Secektion
• Iteratively generates offspring:
for i:=1 to dochoose recombinant r1 uniformly at random
from parent_population;choose recombinant r2 <> r1 uniformly at random
from parent population;offspring := recombine(r1,r2);add offspring to offspring_population;
od
![Page 55: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/55.jpg)
LIACS Natural Computing Group Leiden University55
Operators: Recombination• How does recombination work?
• Discrete recombination:
– Variable at position i will be copied at random (uniformly distr.) from parent 1 or parent 2, position i.Parent 1
Parent 2
Offspring
![Page 56: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/56.jpg)
LIACS Natural Computing Group Leiden University56
Operators: Recombination• Intermediate recombination:
– Variable at position i is arithmetic mean ofParent 1 and Parent 2, position i.
Parent 1
Parent 2
Offspring
1,1rx
1,2rx
2/)( 1,1, 21 rr xx
![Page 57: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/57.jpg)
LIACS Natural Computing Group Leiden University57
Operators: Recombination
Parent 1
Parent 2
Offspring
…Parent
• Global discrete recombination:
– Considers all parents
![Page 58: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/58.jpg)
LIACS Natural Computing Group Leiden University58
Operators: Recombination
Parent 1
Parent 2
Offspring
…Parent
1,1rx
11,
1
iri
x
1,rx
• Global intermediary recombination:
– Considers all parents1,2r
x
![Page 59: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/59.jpg)
LIACS Natural Computing Group Leiden University59
Evolution Strategy
AlgorithmsSelection
![Page 60: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/60.jpg)
LIACS Natural Computing Group Leiden University60
Operators: ()-Selection
• ()-Selection means:
– parents produce offspring by• (Recombination +)
• Mutation
– These individuals will be considered together
– The best out of will be selected („survive“)• Deterministic selection
– This method guarantees monotony• Deteriorations will never be accepted
= Actual solution candidate= New solution candidate
Recombination may be left outMutation always exists!
![Page 61: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/61.jpg)
LIACS Natural Computing Group Leiden University61
Operators: ()-Selection
• ()-Selection means:
– parents produce offspring by• (Recombination +)
• Mutation
– offspring will be considered alone
– The best out of offspring will be selected• Deterministic selection
– The method doesn‘t guarantee monotony• Deteriorations are possible
• The best objective function value in generation t+1 may be worse than the best in generation t.
![Page 62: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/62.jpg)
LIACS Natural Computing Group Leiden University62
Operators: Selection• Example: (2,3)-Selection
• Example: (2+3)-Selection
Parents don‘t survive!Parents don‘t survive …
… but a worse offspring.
… now this offspring survives.
![Page 63: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/63.jpg)
LIACS Natural Computing Group Leiden University63
Exception!
• Possible occurrences of selection– (1+1)-ES: One parent, one offspring, 1/5-Rule– (1,)-ES: One Parent, offspring
• Example: (1,10)-Strategy• One step size / n self-adaptive step sizes • Mutative step size control• Derandomized strategy
– ()-ES: > 1 parents, offspring• Example: (2,15)-Strategy• Includes recombination• Can overcome local optima
– (+)-strategies: elitist strategies
Operators: Selection
![Page 64: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/64.jpg)
LIACS Natural Computing Group Leiden University64
Evolution Strategy:
Self adaptation ofstep sizes
![Page 65: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/65.jpg)
LIACS Natural Computing Group Leiden University65
Self-adaptation
• No deterministic step size control!
• Rather: Evolution of step sizes
– Biology: Repair enzymes, mutator-genes
• Why should this work at all?
– Indirect coupling: step sizes – progress
– Good step sizes improve individuals
– Bad ones make them worse
– This yields an indirect step size selection
![Page 66: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/66.jpg)
LIACS Natural Computing Group Leiden University66
Self-adaptation: Example
• How can we test this at all?
• Need to know optimal step size …
– Only for very simple, convex objective functions
– Here: Sphere model
– Dynamic sphere model• Optimum locations changes occasionally
2*
1
)()( i
n
ii xxxf
: Optimum*x
![Page 67: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/67.jpg)
LIACS Natural Computing Group Leiden University67
Self-adaptation: ExampleObjective function value
… and smallest step size measured in the population
average …
Largest …
According to theoryff optimal step sizes
![Page 68: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/68.jpg)
LIACS Natural Computing Group Leiden University68
Self-adaptation
• Self-adaptation of one step size
– Perfect adaptation
– Learning time for back adaptation proportional n
– Proofs only for convex functions
• Individual step sizes
– Experiments by Schwefel
• Correlated mutations
– Adaptation much slower
![Page 69: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/69.jpg)
LIACS Natural Computing Group Leiden University69
Evolution Strategy:
Derandomization
![Page 70: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/70.jpg)
LIACS Natural Computing Group Leiden University70
Derandomization
• Goals:
– Fast convergence speed
– Fast step size adaptation
– Precise step size adaptation
– Compromise convergence velocity – convergence reliability
• Idea: Realizations of N(0, ) are important!
– Step sizes and realizations can be much different from each other
– Accumulates information over time
![Page 71: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/71.jpg)
LIACS Natural Computing Group Leiden University71
Derandomzed (1,)-ES
kg
scalggg
N Zxxk
• Current parent: in generation g
• Mutation (k=1,…,): Offspring k
Global step size in generation g
Individual step sizes in generation g
gx
)1,0(~),...,( 1 NzzzZ in
• Selection: Choice of best offspring
gN
g
selxx 1 Best of offspring
in generation g
![Page 72: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/72.jpg)
LIACS Natural Computing Group Leiden University72
Derandomized (1,)-ES
selgA
gA ZcZcZ
1)1(
• Accumulation of selected mutations:
The particular mutation vector,which created the parent!
• Also: weighted history of good mutation vectors!• Initialization:
• Weight factor:00
AZ
nc
1
![Page 73: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/73.jpg)
LIACS Natural Computing Group Leiden University73
Derandomized (1,)-ES
n
cc
n
Z gAgg
5
11
2
exp1
• Step size adaptation: Norm of vector
Vector of absolute values
scal
cc
Z gAg
scalg
scal
35.0
2
1 Regulates adaptation
speed and precision
![Page 74: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/74.jpg)
LIACS Natural Computing Group Leiden University74
Derandomized (1,)-ES• Explanations:
– Normalization of average variations in case of missing selection (no bias):
– Correction for small n: 1/(5n)
– Learning rates:
c
c
2
n
n
scal /1
/1
![Page 75: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/75.jpg)
LIACS Natural Computing Group Leiden University75
Evolution Strategy:
Rules of thumb
![Page 76: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/76.jpg)
LIACS Natural Computing Group Leiden University76
Some Theory Highlights
• Convergence velocity:
• For (1,)-strategies:
• For ()-strategies (discrete and intermediary recombination):
n/1~
Problem dimensionality
ln~
Speedup by is just logarithmic – more processors are only to a limited extend useful to increase
ln~ Genetic Repair Effect
of recombination!
Genetic Repair Effectof recombination!
![Page 77: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/77.jpg)
LIACS Natural Computing Group Leiden University77
…
• For strategies with global intermediary recombination:
• Good heuristic for (1,):
2/
log34
n
9.5219.03150
9.4118.82140
9.3018.60130
9.1818.36120
9.0518.10110
8.9117.82100
8.7517.5090
8.5717.1580
8.3716.7570
8.1416.2860
7.8715.7450
7.5315.0740
7.1014.2030
6.4912.9920
5.4510.9110
n
10
![Page 78: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/78.jpg)
LIACS Natural Computing Group Leiden University
Mixed-IntegerEvolution Strategies
![Page 79: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/79.jpg)
LIACS Natural Computing Group Leiden University79
Mixed-Integer Evolution Strategy
• Generalized optimization problem:
![Page 80: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/80.jpg)
LIACS Natural Computing Group Leiden University80
Mixed-Integer ES: Mutation
Learning rates (global)
Learning rates (global)
Geometrical distribution
Mutation Probabilities
![Page 81: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/81.jpg)
LIACS Natural Computing Group Leiden University
Some Application Examples
![Page 82: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/82.jpg)
LIACS Natural Computing Group Leiden University82
Safety Optimization – Pilot Study
• Aim: Identification of most appropriate Optimization Algorithm for realistic example!
• Optimizations for 3 test cases and 14 algorithms were performed (28 x 10 = 280 shots)
– Body MDO Crash / Statics / Dynamics– MCO B-Pillar– MCO Shape of Engine Mount
• NuTech’s ES performed significantly better than Monte-Carlo-scheme, GA, and Simulated Annealing
• Results confirmed by statistical hypothesis testing
![Page 83: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/83.jpg)
LIACS Natural Computing Group Leiden University83
MDO Crash / Statics / Dynamics
• Minimization of body mass • Finite element mesh
– Crash ~ 130.000 elements– NVH ~ 90.000 elements
• Independent parameters: Thickness of each unit: 109
• Constraints: 18
Algorithm Avg. reduction (kg) Max. reduction (kg) Min. reduction (kg)
Best so far -6.6 -8.3 -3.3
NuTech ES -9.0 -13.4 -6.3
![Page 84: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/84.jpg)
LIACS Natural Computing Group Leiden University84
MCO B-Pillar – Side Crash
• Minimization of mass & displacement
• Finite element mesh– ~ 300.000 elements
• Independent parameters: Thickness of 10 units
• Constraints: 0
• ES successfully developed Pareto front
Mass
Intr
usio
n
![Page 85: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/85.jpg)
LIACS Natural Computing Group Leiden University85
MCO Shape of Engine Mount
• Mass minimal shape withaxial load > 90 kN
• Finite element mesh– ~ 5000 elements
• Independent parameters:9 geometry variables
• Dependent parameters: 7• Constraints: 3• ES optimized mount
– less weight than mount optimized with best so far method
– geometrically better deformation
![Page 86: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/86.jpg)
LIACS Natural Computing Group Leiden University86
Safety Optimization – Example of use
• Production Run !• Minimization of body mass • Finite element mesh
– Crash ~ 1.000.000 elements– NVH ~ 300.000 elements
• Independent parameters: – Thickness of each unit: 136
• Constraints: 47, resulting from various loading cases• 180 (10 x 18) shots ~ 12 days• No statistical evaluation due to problem complexity
![Page 87: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/87.jpg)
LIACS Natural Computing Group Leiden University87
Safety Optimization – Example of use
• 13,5 kg weight reduction by NuTech’s ES• ~ 2 kg more mass reduction than Best so far method• Typically higher convergence velocity of ES
~ 45% less time (~ 3 days saving) for comparable quality needed• Still potential of improvements after 180 shots.• Reduction of development time from 5 to 2 weeks allows for process
integration
Best so far Method
Generations
Mas
s
Initial Value
Generations
Mas
s
Initial Value
NuTech’s Evolution Strategy
![Page 88: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/88.jpg)
LIACS Natural Computing Group Leiden University88
MDO – Optimization Run
107,13%
105,05%
100,00%
All Individuals
Number of Individuals
Fitness Infeasible
Feasible
![Page 89: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/89.jpg)
LIACS Natural Computing Group Leiden University89
Optimization of metal stamping process
• Objective: Minimization of defects in the produced parts.
• Variables: Geometric parameters and forces.
• ES finds very good results in short time
• Computationally expensive simulation
![Page 90: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/90.jpg)
LIACS Natural Computing Group Leiden University90
Bridgeman Casting Process
• Objective: Max. homogeneity of workpiece after Casting Process
• Variables: 18 continuous speed variables for Casting Schedule
• Computationally expensive simulation (up to 32h simulation time)
Initial (DoE) GCM (Commercial Gradient Based Method)
Evolution Strategy
Turbine Bladeafter Casting
![Page 91: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/91.jpg)
LIACS Natural Computing Group Leiden University91
Algorithm 100 OFE
200 OFE
1000OFE
SQP 148 67.7 59
Conjugate directions
135 135 135
Pattern search
157 147 48
DSCP/Rosenbrock
88 38 29
Complex strategy
152 152 111
(1,3)-DES 91 67 67
(1,5)-DES 53 48 48
(1,7)-DES 110 58 58
(1,10)-DES 152 80 28
MAES 122 65 12
Minimize thedeviation of temperture atthe cube´s Surface to1000°C !
maxT
Input: TemperatureProfile (12 Variables7 Temperatures and 5 Time-Steps)
Steel Cube Temperature Control
![Page 92: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/92.jpg)
LIACS Natural Computing Group Leiden University92
- Maximisation of growth speed- Minimisation of diameter differences
FLUENT: Simulation of fluid flowCASTS: Calculation of Temperatureand concentration field
0 m/s 3 m/s1.5 m/s
Reaction Gas TCS Growing High Purity Silicon Rod
Optimsation of 15 process parameters:
Production time 35 hours 30 hours
Siemens C Reactor
![Page 93: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/93.jpg)
LIACS Natural Computing Group Leiden University93
Multipoint Airfoil Optimiziation• Objective: Find pressure profiles that are a compromise between two given
target pressure distributions under two given flow conditions!
• Variables: 12 to 18 Bezier points for the airfoil
High Lift!
Low Drag!
Start
Cruise
![Page 94: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/94.jpg)
LIACS Natural Computing Group Leiden University94
Traffic Light Control Optimization
• Objective: Minimization of total delay / number of stops• Variables: Green times for next switching schedule• Dynamic optimization, depending on actual traffic • Better results (3-5%)• Higher flexibility than with traditional controllers
![Page 95: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/95.jpg)
LIACS Natural Computing Group Leiden University95
Optimization of elevator control
• Minimization of passenger waiting times.• Better results (~10%) than with traditional controller.• Dynamic optimization, depending on actual traffic.
![Page 96: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/96.jpg)
LIACS Natural Computing Group Leiden University96
Optimization of network routing
• Minimization of end-to-end-blockings under service constraints.
• Optimization of routing tables for existing, hard-wired networks.
• 10%-1000% improvement.
![Page 97: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/97.jpg)
LIACS Natural Computing Group Leiden University97
Automatic battery configuration
• Configuration and optimization of industry batteries.
• Based on user specifications, given to the system.
• Internet-Configurator (Baan, Hawker, NuTech).
![Page 98: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/98.jpg)
LIACS Natural Computing Group Leiden University98
Optimization of reactor fueling.
• Minimization of total costs.
• Creates new fuel assembly reload patterns.
• Clear improvements (1%-5%)of existing expert solutions.
• Huge cost saving.
![Page 99: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/99.jpg)
LIACS Natural Computing Group Leiden University99
Optical Coatings: Design Optimization
• Nonlinear mixed-integer problem, variable dimensionality.
• Minimize deviation from desired reflection behaviour.
• Excellent synthesis method; robust and reliable results.
• MOC-Problems: anti-reflection coating, dialetric filters
• Robust design: Sharp peaks vs. Robust peaks
![Page 100: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/100.jpg)
LIACS Natural Computing Group Leiden University100
Example: IntravascularUltrasound Image Analysis
• Real-time high-resolution tomographic images from the inside of coronary vessels:
![Page 101: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/101.jpg)
LIACS Natural Computing Group Leiden University101
Intravascular Ultrasound Image Analysis
• Detected Features in an IVUS Image:
Vessel Border
Sidebranch
Lumen
Shadow
Plague
![Page 102: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/102.jpg)
LIACS Natural Computing Group Leiden University102
Experimental Results on IVUS images• Parameters for the lumen feature detector
![Page 103: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/103.jpg)
LIACS Natural Computing Group Leiden University103
Intravascular Ultrasound Image Analysis: Results
• On each of 5 data sets algorithm ran for 2804 evaluations – 19,5h of total computing time
• Significant improvement over expert tuning
Performance of the best found MI-ES parameter solutions
• A paired two-tailed t-test was performed on the difference measurements for each image dataset using a 95% confidence interval (p=0.05)
• The null-hypothesis is that the mean difference results of the best MI-ES individual and the default parameters are equal.
![Page 104: LIACS Natural Computing Group Leiden University Evolutionary Algorithms](https://reader035.vdocument.in/reader035/viewer/2022062518/56649c875503460f9493e4b2/html5/thumbnails/104.jpg)
LIACS Natural Computing Group Leiden University104
Literature
• H.-P. Schwefel: Evolution and Optimum Seeking, Wiley, NY, 1995.
• I. Rechenberg: Evolutionsstrategie 94, frommann-holzboog, Stuttgart, 1994.
• Th. Bäck: Evolutionary Algorithms in Theory and Practice, Oxford University Press, NY, 1996.
• Th. Bäck, D.B. Fogel, Z. Michalewicz (Hrsg.): Handbook of Evolutionary Computation, Vols. 1,2, Institute of Physics Publishing, 2000.