Cultural Algorithms

Оглавление

Robert G. Reynolds. Cultural Algorithms

Table of Contents

List of Tables

List of Illustrations

Guide

Pages

Cultural Algorithms. Tools to Model Complex Dynamic Social Systems

List of Contributors

About the Companion Website

1 System Design Using Cultural Algorithms

Introduction

The Cultural Engine

Zeroth Law of Thermodynamics, About Thermal Equilibrium

First Law of Thermodynamics, About the Conservation of Energy

Second Law of Thermodynamics, About Entropy

Third Law of Thermodynamics, About the Absolute Zero of Temperature

Outline of the Book: Cultural Learning in Dynamic Environments

References

Note

2 The Cultural Algorithm Toolkit System

CAT Overview

Downloading and Running CAT

The Repast Simphony System

Knowledge Sources

Fitness Functions

ConesWorld

The Logistics Function

CAT Sample Runs: ConesWorld

CAT Sample Runs: Other Problems

Reference

Note

3 Social Learning in Cultural Algorithms with Auctions

Introduction

Cultural Algorithms

Subcultured Multi‐Layered, Deep Heterogeneous Networks

Auction Mechanisms

The Cultural Engine

ConesWorld

Experimental Framework

Results

Conclusions

References

Note

4 Using Common Value Auction in Cultural Algorithm to Enhance Robustness and Resilience of Social Knowledge Distribution Systems

Cultural Algorithms

Common Value Auction

ConesWorld

Dynamic Experimental Framework

Results

Conclusions and Future Work

References

Note

5 Optimizing AI Pipelines: A Game‐Theoretic Cultural Algorithms Approach

Introduction

Overview of Cultural Algorithms

CA Knowledge Distribution Mechanisms

Primer on Game Theory

Game‐Theoretic Knowledge Distribution

Continuous‐Action Iterated Prisoner's Dilemma

Play

Payoff

Outcome

Case 1

Case 2

Case 3

Learning Rate Adjustment

Test Results: Benchmark Problem

Test Results: Computer Vision Pipeline

Conclusions

References

Note

6 Cultural Algorithms for Social Network Analysis: Case Studies in Team Formation

Introduction

Application of Social Network

Forming Successful Teams

Formulating TFP

Definition: (Team of Experts)

Communication Cost

Definition: (Sum of Distances)

Definition: (Diameter)

Personnel Cost

Definition: (Sum of Personnel Cost)

Distance Cost

Definition: (Distance Cost)

Workload Balance

Why Artificial Intelligence?

Cultural Algorithms

Forming Teams in Coauthorship Network

Individual Representation

Fitness Function

Belief Space

Dataset and Observations

Skill Frequency

Forming Teams in Health‐care Network

Individual Representation

Fitness Function

Dataset and Observation

Summary and Conclusion

References

Notes

7 Evolving Emergent Team Strategies in Robotic Soccer using Enhanced Cultural Algorithms

Introduction

Related Work

The 2D Soccer Simulation Test Bed. Prototyping Environment Motivation

Simulator Overview

Teams

Evolution of Team Strategies via Cultural Algorithm. Basic Engine of CA

Knowledge Sources in CA

Evolution of Team Strategies

Modified Cultural Algorithm. Encoding

Fitness Function

Enhancing the Defense

Experiments and Analysis of Results

Experiment 1. Enhanced CA Team Against the Default Team

Experiment 2. Enhanced CA Team Against the Team with a Strong Offense

Experiment 3. Enhanced CA Team Against the Team with Enhanced Defense

Experiment 4. Enhanced CA Team Against the GA Team

Conclusion

References

8 The Use of Cultural Algorithms to Learn the Impact of Climate on Local Fishing Behavior in Cerro Azul, Peru

Introduction

An Overview of the Cerro Azul Fishing Dataset. Introduction

An Overview of the Database Content as a Complex System

Data Mining at the Macro, Meso, and Micro Levels. Cerro Azul can be Viewed as a Complex System

The Three Levels of Data Analysis

Cultural Algorithms and Multiobjective Optimization. Introduction

Introduction to the Cultural Algorithm

CAPSO, Cultural Algorithm, and Particle Swarm Optimizer

The Artisanal Fishing Model. Biobjective

The Agent‐Based Model

The Trip_Graph Model

Nondominance

The Experimental Results. Introduction

Full Week, Pareto Frontier

No Sundays, Pareto Frontier

Monday, Tuesday, Wednesday, Pareto Frontier

Thursday, Friday, Saturday, Pareto Frontier

Summary of Experimental Results

Statistical Validation. Overall Performance Comparison

Hypothesis. Null Hypothesis (H0)

Summary

Conclusions and Future Work. Conclusions

Future Work

References

Note

9 CAPSO: A Parallelized Multiobjective Cultural Algorithm Particle Swarm Optimizer

Introduction

Multiobjective Optimization. Overview

Formulating a Multiobjective Problem

Cultural Algorithms. Background

Overview

Acceptance Step

Belief Space Update Step

Influence Step

Knowledge Sources

CAPSO Knowledge Structures

Situational Knowledge

Normative Knowledge

Historical Knowledge

Domain Knowledge

Topographic Knowledge

Tracking Knowledge Source Progress (Other than Topographic)

CAPSO Algorithm Pseudocode

Multiple Runs

Comparison of Benchmark Problems

CONSTR

SRN

TNK

KITA

Overall Summary of Results

Other Applications

References

Note

10 Exploring Virtual Worlds with Cultural Algorithms: Ancient Alpena–Amberley Land Bridge

Archaeological Challenges

Generalized Framework

The Land Bridge Hypothesis

Origin and Form

Putting Data to Work

Pathfinding and Planning

Identifying Good Locations: The Hotspot Finder

Cultural Algorithms

Cultural Algorithm Mechanisms

The Composition of the Belief Space

Future Work

Path Planning Strategy

Local Tactics

Detailed Locational Information

Extending the CA

Human Presence in the Virtual World

Increasing the Complexity

Updated Path‐Planning Results in Unity

The Fully Rendered Land Bridge

Pathfinder Mechanisms

Results

Conclusions

References

Note

Index. a

b

c

d

e

f

g

h

i

k

l

m

n

o

p

r

s

t

u

v

w

z

IEEE Press Series on COMPUTATIONAL INTELLIGENCE

WILEY END USER LICENSE AGREEMENT

Отрывок из книги

IEEE Press 445 Hoes Lane Piscataway, NJ 08854

.....

In addition to the constraints and basic domain knowledge of the ConesWorld problem, there is also a series of flags for each cone. These flags indicate the directional change a cone may take during an update. There are strict maximum and minimum values that the dimension variables of a cone may change between. The movement between these extremes is cyclical in nature, moving the cone first toward one extreme and then back toward another, similar to how the pistons in an engine can move up and down. Should a cone undergoing an update exceed its dimensional limits, its flag is switched to denote its new increasing or decreasing status, and the amount by which it exceeding its limit then becomes additional change in its new dimension.

For an example of this, assume the cone height limits are set between 1 and 3, and the rate of change is restricted to 0.25 at the most for any single update. If a cone's height is already at 2.80 with its directional flag indicating that it should increase, it will have 0.25 added to it, and it will exceed the height limit of 3 by 0.05. In this case, its flag will be triggered to indicate that it is now decreasing, and the overflow of 0.05 will be taken away from the height of the cone, resulting in a cone of 2.95 height. The next update to the cone, for example 0.20, will result in the cone's height being updated to 2.75, as its directional flag is still indicating that it is decreasing. This decrease will continue until the cone hits the lower limit, at which point the flag will be toggled, the overflow will be added back in, and the cycle will continue again.

.....