SlideShare ist ein Scribd-Unternehmen logo
1 von 20
An Analysis of a Selecto-Lamarckian Model of
Multimemetic Algorithms with Dynamic
Self-Organized Topology
Rafael Nogueras1
Juan L.J. Laredo3

Carlos Cotta1 Carlos M. Fernandes2
Juan J. Merelo4 Agostinho C. Rosa2

1 Universidad
3 University

de M´laga (Spain), 2 Technical University Lisbon (Portugal),
a
of Luxembourg (Luxembourg), 4 University of Granada (Spain)

TPNC 2013, C´ceres, 3-5 December 2013
a

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
What are Memes?
Memes are information pieces that constitute units of imitation.
“Examples of memes are tunes, ideas,
catch-phrases, clothes fashions, ways of
making pots or of building arches. Just as
genes propagate themselves in the gene pool
by leaping from body to body via sperms or
eggs, so memes propagate themselves in the
meme pool by leaping from brain to brain via a
process which, in the broad sense, can be
called imitation.”
The Selfish Gene, Richard Dawkins, 1976

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
What is a Memetic Algorithm?

Memetic Algorithms
A Memetic Algorithm is a population of agents
that alternate periods of self-improvement
with periods of cooperation, and competition.
Pablo Moscato, 1989
Memes can be implicitly defined be the choice of local-search (i.e.,
self-improvement) method, or can be explicitly described in the
agent.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Multimemetic Algoritms (and Memetic Computing)
The term “multimemetic” was coined by N. Krasnogor and J.
Smith (2001). In a MMA, each agent carries a solution and the
meme(s) to improve it.
Evolution works at these two levels, cf. Moscato (1999).
Memetic Computing
A paradigm that uses the notion of meme(s) as units of
information encoded in computational representations for the
purpose of problem solving.
Ong, Lim, Chen, 2010

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Scope

Some interesting issues in MMAs:
Memes evolve in MMAs alongside with the solutions they
attach to. It is up to the algorithm to (self-adaptively)
discover good fits between genotypes and memes.
Memes are indirectly assessed via the effect they have on
genotypes.
We consider an analyze an idealized model of MMAs to analyze
meme propagation with dynamic self-organized spatial structures.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Background

Dynamic of meme propagation is more complex than genetic
counterparts.
Genes represent solutions objectively measurable via the
fitness function.
Memes are indirectly evaluated by their effect on solutions.
A first analysis was done by Nogueras and Cotta (2013) with
panmictic and spatially-structured populations. Population
structure is very important to determine the behavior of the
algorithm.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Dynamic Self-Organized Topology

A dynamic model defined by Fernandes et al. (2012) combining
ideas from swarm intelligence and cellular automata is considered
in this study.
Model uses simple rules for movement on a
large 2D-lattice, giving rise to self-organized
clusters of particles.
The clusters evolve and change their shape
with some kind of dynamic order.
We consider how memes propagate in this environment.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Preliminaries
Each agent is a pair g , m ∈ R2 – i.e., gene, meme . The effect
of a meme is captured by a function f : R2 → R, i.e.,
meme application

g , m − − − − − → f (g , m), m
−−−−−
m actually represents the improvement potential of the meme.
lim f n (g , m) = m if g < m

n→∞

f (g , m) = g

if g

m

The population P = [ g1 , m1 , · · · , gµ , mµ ] of the MMA is a
collection of µ such agents.
Agent communication is constrained by a spatial structure,
characterized by a µ × µ Boolean matrix S.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Model Pseudocode

Algorithm 1: Selecto-Lamarckian Model
for i ∈ [1 · · · µ] do
Initialize gi , mi ;
end
while ¬ Converged (P) do
i ← URand(1, µ) // Pick random location
g , m ←Selection(P, S, i);
g ← f (g , m) // Local improvement
P ← Replace(P, S, i, g , m );
end

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Concepts

Let G be a grid of size r × s > µ. Each cell Guv of the grid is a
tuple (ηuv , ζuv ), where:
ηuv ∈ {1, · · · , µ} ∪ {•} and ζuv ∈ (D × N) ∪ {•}.
ηuv indicates the index of the individual that occupies position
u, v in the grid.
ζuv is a mark placed by individuals which occupied that
position in the past, where:
f
ζuv is the fitness value of the individual.
t
ζuv is a time stamp.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Individual Movement
The system combine ideas from swarm intelligence and cellular
automata.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Individual Movement

Algorithm 2: Individual Movement (i, t, G )
u, v ← ρ(i); move←true;
f
if exists u (i) , v (i) ∈ N u, v such that ζuv > gi then
f
f
u , v ← arg min{ζuv | ζuv > gi };
else
f
if exists u (i) , v (i) ∈ N u, v such that ζuv < gi then
f
f
u , v ← arg max{ζuv | ζuv < gi };
else
if N u, v = ∅ then
Pick u , v at random from N u, v ;
else
move ← false;
end
end
end
if move then
f
t
ζuv ← gi ; ζuv ← t; // mark old cell
ηuv = •; ηu v = i; // move to new cell
end

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Setting
Goal
Explore the dynamics of meme propagation and how it is affected
by factors such as the selection probability, the improvement
potential of memes and the spatial structure of the population.
µ = 256.
pLS ∈ {1/256, 0.1, 0.5, 1.0}.
Spatial structure:
1
2
3

Panmictic: full connectivity with static structure.
Von Neumman neighborhood (r = 1) with static structure.
Moore neighborhood (r = 1) with dynamic structure.

Meme application:
f (g , m) =

g
(g + m)/2

R. Nogueras et al.

if g m
if g < m

MMAs with Dynamic Self-Organized Topology
Numerical Simulations
Individual Distribution – Evolution

Individuals start from a random distribution and quickly group in
clusters during the run.
R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Numerical Simulations
Growth Curves

The number of copies of the dominant meme grows until taking
over the population:
the panmictic model is the first to converge.
the dynamic model is closer to von Neumann model
depending on the value of pS .
R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Numerical Simulations
Qualified Run-Time Distributions – α = 1

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Numerical Simulations
Qualified Run-Time Distributions – α = 1/2

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Numerical Simulations
Spectral Analysis

The spectrum of the average number of neighbors indicates:
the intensity is proportional to f a for some a < 0.
the spectrum slope is closer to pink noise.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Conclusions

A dynamic model provides promising results in comparison to
unstructured populations and to populations arranged in static
lattices.
By tuning the ratio between self-organization and evolution the
convergence of the algorithm can be adjusted.
Future work:
other topologies and movement policies,
decouple movement for neighborhood and evolutionary
interaction,
full-fledged MMA.

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology
Thank You!

Please find us in Facebook
http://facebook.com/AnySelfProject
and in Twitter
@anyselfproject
AnySelf Project

R. Nogueras et al.

MMAs with Dynamic Self-Organized Topology

Weitere ähnliche Inhalte

Ähnlich wie An Analysis of a Selecto-Lamarckian Model of Multimemetic Algorithms with Dynamic Self-Organized Topology

A comparison of classification techniques for seismic facies recognition
A comparison of classification techniques for seismic facies recognitionA comparison of classification techniques for seismic facies recognition
A comparison of classification techniques for seismic facies recognition
Pioneer Natural Resources
 
Be2419772016
Be2419772016Be2419772016
Be2419772016
IJMER
 
A Neural Probabilistic Language Model.pptx
A Neural Probabilistic Language Model.pptxA Neural Probabilistic Language Model.pptx
A Neural Probabilistic Language Model.pptx
Rama Irsheidat
 
ABC & Empirical Lkd
ABC & Empirical LkdABC & Empirical Lkd
ABC & Empirical Lkd
Deb Roy
 
Dowload Paper.doc.doc
Dowload Paper.doc.docDowload Paper.doc.doc
Dowload Paper.doc.doc
butest
 

Ähnlich wie An Analysis of a Selecto-Lamarckian Model of Multimemetic Algorithms with Dynamic Self-Organized Topology (20)

NEURAL MODEL-APPLYING NETWORK (NEUMAN): A NEW BASIS FOR COMPUTATIONAL COGNITION
NEURAL MODEL-APPLYING NETWORK (NEUMAN): A NEW BASIS FOR COMPUTATIONAL COGNITIONNEURAL MODEL-APPLYING NETWORK (NEUMAN): A NEW BASIS FOR COMPUTATIONAL COGNITION
NEURAL MODEL-APPLYING NETWORK (NEUMAN): A NEW BASIS FOR COMPUTATIONAL COGNITION
 
F043046054
F043046054F043046054
F043046054
 
F043046054
F043046054F043046054
F043046054
 
F043046054
F043046054F043046054
F043046054
 
Speech Recognition using HMM & GMM Models: A Review on Techniques and Approaches
Speech Recognition using HMM & GMM Models: A Review on Techniques and ApproachesSpeech Recognition using HMM & GMM Models: A Review on Techniques and Approaches
Speech Recognition using HMM & GMM Models: A Review on Techniques and Approaches
 
Energy-Based Models with Applications to Speech and Language Processing
Energy-Based Models with Applications to Speech and Language ProcessingEnergy-Based Models with Applications to Speech and Language Processing
Energy-Based Models with Applications to Speech and Language Processing
 
A comparison of classification techniques for seismic facies recognition
A comparison of classification techniques for seismic facies recognitionA comparison of classification techniques for seismic facies recognition
A comparison of classification techniques for seismic facies recognition
 
Be2419772016
Be2419772016Be2419772016
Be2419772016
 
Genetic algorithm
Genetic algorithmGenetic algorithm
Genetic algorithm
 
SWARM INTELLIGENCE FROM NATURAL TO ARTIFICIAL SYSTEMS: ANT COLONY OPTIMIZATION
SWARM INTELLIGENCE FROM NATURAL TO ARTIFICIAL SYSTEMS: ANT COLONY OPTIMIZATIONSWARM INTELLIGENCE FROM NATURAL TO ARTIFICIAL SYSTEMS: ANT COLONY OPTIMIZATION
SWARM INTELLIGENCE FROM NATURAL TO ARTIFICIAL SYSTEMS: ANT COLONY OPTIMIZATION
 
A COMPREHENSIVE SURVEY OF GREY WOLF OPTIMIZER ALGORITHM AND ITS APPLICATION
A COMPREHENSIVE SURVEY OF GREY WOLF OPTIMIZER ALGORITHM AND ITS APPLICATIONA COMPREHENSIVE SURVEY OF GREY WOLF OPTIMIZER ALGORITHM AND ITS APPLICATION
A COMPREHENSIVE SURVEY OF GREY WOLF OPTIMIZER ALGORITHM AND ITS APPLICATION
 
AI/ML session by GDSC ZHCET AMU, ALIGARH
AI/ML session by GDSC ZHCET AMU, ALIGARHAI/ML session by GDSC ZHCET AMU, ALIGARH
AI/ML session by GDSC ZHCET AMU, ALIGARH
 
(CoSECiVi'14) Evolving Evil: Optimizing Flocking Strategies through Genetic A...
(CoSECiVi'14) Evolving Evil: Optimizing Flocking Strategies through Genetic A...(CoSECiVi'14) Evolving Evil: Optimizing Flocking Strategies through Genetic A...
(CoSECiVi'14) Evolving Evil: Optimizing Flocking Strategies through Genetic A...
 
Swarm intel
Swarm intelSwarm intel
Swarm intel
 
A Neural Probabilistic Language Model.pptx
A Neural Probabilistic Language Model.pptxA Neural Probabilistic Language Model.pptx
A Neural Probabilistic Language Model.pptx
 
Behavior study of entropy in a digital image through an iterative algorithm
Behavior study of entropy in a digital image through an iterative algorithmBehavior study of entropy in a digital image through an iterative algorithm
Behavior study of entropy in a digital image through an iterative algorithm
 
ABC in Venezia
ABC in VeneziaABC in Venezia
ABC in Venezia
 
ABC & Empirical Lkd
ABC & Empirical LkdABC & Empirical Lkd
ABC & Empirical Lkd
 
ABC and empirical likelihood
ABC and empirical likelihoodABC and empirical likelihood
ABC and empirical likelihood
 
Dowload Paper.doc.doc
Dowload Paper.doc.docDowload Paper.doc.doc
Dowload Paper.doc.doc
 

Kürzlich hochgeladen

Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Victor Rentea
 

Kürzlich hochgeladen (20)

Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..Understanding the FAA Part 107 License ..
Understanding the FAA Part 107 License ..
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
WSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering DevelopersWSO2's API Vision: Unifying Control, Empowering Developers
WSO2's API Vision: Unifying Control, Empowering Developers
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWEREMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
EMPOWERMENT TECHNOLOGY GRADE 11 QUARTER 2 REVIEWER
 
MS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectorsMS Copilot expands with MS Graph connectors
MS Copilot expands with MS Graph connectors
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
Exploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with MilvusExploring Multimodal Embeddings with Milvus
Exploring Multimodal Embeddings with Milvus
 
MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024Finding Java's Hidden Performance Traps @ DevoxxUK 2024
Finding Java's Hidden Performance Traps @ DevoxxUK 2024
 
Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdfRising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
Rising Above_ Dubai Floods and the Fortitude of Dubai International Airport.pdf
 
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 AmsterdamDEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
DEV meet-up UiPath Document Understanding May 7 2024 Amsterdam
 
AWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of TerraformAWS Community Day CPH - Three problems of Terraform
AWS Community Day CPH - Three problems of Terraform
 
Platformless Horizons for Digital Adaptability
Platformless Horizons for Digital AdaptabilityPlatformless Horizons for Digital Adaptability
Platformless Horizons for Digital Adaptability
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 

An Analysis of a Selecto-Lamarckian Model of Multimemetic Algorithms with Dynamic Self-Organized Topology

  • 1. An Analysis of a Selecto-Lamarckian Model of Multimemetic Algorithms with Dynamic Self-Organized Topology Rafael Nogueras1 Juan L.J. Laredo3 Carlos Cotta1 Carlos M. Fernandes2 Juan J. Merelo4 Agostinho C. Rosa2 1 Universidad 3 University de M´laga (Spain), 2 Technical University Lisbon (Portugal), a of Luxembourg (Luxembourg), 4 University of Granada (Spain) TPNC 2013, C´ceres, 3-5 December 2013 a R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 2. What are Memes? Memes are information pieces that constitute units of imitation. “Examples of memes are tunes, ideas, catch-phrases, clothes fashions, ways of making pots or of building arches. Just as genes propagate themselves in the gene pool by leaping from body to body via sperms or eggs, so memes propagate themselves in the meme pool by leaping from brain to brain via a process which, in the broad sense, can be called imitation.” The Selfish Gene, Richard Dawkins, 1976 R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 3. What is a Memetic Algorithm? Memetic Algorithms A Memetic Algorithm is a population of agents that alternate periods of self-improvement with periods of cooperation, and competition. Pablo Moscato, 1989 Memes can be implicitly defined be the choice of local-search (i.e., self-improvement) method, or can be explicitly described in the agent. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 4. Multimemetic Algoritms (and Memetic Computing) The term “multimemetic” was coined by N. Krasnogor and J. Smith (2001). In a MMA, each agent carries a solution and the meme(s) to improve it. Evolution works at these two levels, cf. Moscato (1999). Memetic Computing A paradigm that uses the notion of meme(s) as units of information encoded in computational representations for the purpose of problem solving. Ong, Lim, Chen, 2010 R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 5. Scope Some interesting issues in MMAs: Memes evolve in MMAs alongside with the solutions they attach to. It is up to the algorithm to (self-adaptively) discover good fits between genotypes and memes. Memes are indirectly assessed via the effect they have on genotypes. We consider an analyze an idealized model of MMAs to analyze meme propagation with dynamic self-organized spatial structures. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 6. Background Dynamic of meme propagation is more complex than genetic counterparts. Genes represent solutions objectively measurable via the fitness function. Memes are indirectly evaluated by their effect on solutions. A first analysis was done by Nogueras and Cotta (2013) with panmictic and spatially-structured populations. Population structure is very important to determine the behavior of the algorithm. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 7. Dynamic Self-Organized Topology A dynamic model defined by Fernandes et al. (2012) combining ideas from swarm intelligence and cellular automata is considered in this study. Model uses simple rules for movement on a large 2D-lattice, giving rise to self-organized clusters of particles. The clusters evolve and change their shape with some kind of dynamic order. We consider how memes propagate in this environment. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 8. Preliminaries Each agent is a pair g , m ∈ R2 – i.e., gene, meme . The effect of a meme is captured by a function f : R2 → R, i.e., meme application g , m − − − − − → f (g , m), m −−−−− m actually represents the improvement potential of the meme. lim f n (g , m) = m if g < m n→∞ f (g , m) = g if g m The population P = [ g1 , m1 , · · · , gµ , mµ ] of the MMA is a collection of µ such agents. Agent communication is constrained by a spatial structure, characterized by a µ × µ Boolean matrix S. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 9. Model Pseudocode Algorithm 1: Selecto-Lamarckian Model for i ∈ [1 · · · µ] do Initialize gi , mi ; end while ¬ Converged (P) do i ← URand(1, µ) // Pick random location g , m ←Selection(P, S, i); g ← f (g , m) // Local improvement P ← Replace(P, S, i, g , m ); end R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 10. Concepts Let G be a grid of size r × s > µ. Each cell Guv of the grid is a tuple (ηuv , ζuv ), where: ηuv ∈ {1, · · · , µ} ∪ {•} and ζuv ∈ (D × N) ∪ {•}. ηuv indicates the index of the individual that occupies position u, v in the grid. ζuv is a mark placed by individuals which occupied that position in the past, where: f ζuv is the fitness value of the individual. t ζuv is a time stamp. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 11. Individual Movement The system combine ideas from swarm intelligence and cellular automata. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 12. Individual Movement Algorithm 2: Individual Movement (i, t, G ) u, v ← ρ(i); move←true; f if exists u (i) , v (i) ∈ N u, v such that ζuv > gi then f f u , v ← arg min{ζuv | ζuv > gi }; else f if exists u (i) , v (i) ∈ N u, v such that ζuv < gi then f f u , v ← arg max{ζuv | ζuv < gi }; else if N u, v = ∅ then Pick u , v at random from N u, v ; else move ← false; end end end if move then f t ζuv ← gi ; ζuv ← t; // mark old cell ηuv = •; ηu v = i; // move to new cell end R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 13. Setting Goal Explore the dynamics of meme propagation and how it is affected by factors such as the selection probability, the improvement potential of memes and the spatial structure of the population. µ = 256. pLS ∈ {1/256, 0.1, 0.5, 1.0}. Spatial structure: 1 2 3 Panmictic: full connectivity with static structure. Von Neumman neighborhood (r = 1) with static structure. Moore neighborhood (r = 1) with dynamic structure. Meme application: f (g , m) = g (g + m)/2 R. Nogueras et al. if g m if g < m MMAs with Dynamic Self-Organized Topology
  • 14. Numerical Simulations Individual Distribution – Evolution Individuals start from a random distribution and quickly group in clusters during the run. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 15. Numerical Simulations Growth Curves The number of copies of the dominant meme grows until taking over the population: the panmictic model is the first to converge. the dynamic model is closer to von Neumann model depending on the value of pS . R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 16. Numerical Simulations Qualified Run-Time Distributions – α = 1 R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 17. Numerical Simulations Qualified Run-Time Distributions – α = 1/2 R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 18. Numerical Simulations Spectral Analysis The spectrum of the average number of neighbors indicates: the intensity is proportional to f a for some a < 0. the spectrum slope is closer to pink noise. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 19. Conclusions A dynamic model provides promising results in comparison to unstructured populations and to populations arranged in static lattices. By tuning the ratio between self-organization and evolution the convergence of the algorithm can be adjusted. Future work: other topologies and movement policies, decouple movement for neighborhood and evolutionary interaction, full-fledged MMA. R. Nogueras et al. MMAs with Dynamic Self-Organized Topology
  • 20. Thank You! Please find us in Facebook http://facebook.com/AnySelfProject and in Twitter @anyselfproject AnySelf Project R. Nogueras et al. MMAs with Dynamic Self-Organized Topology