Genetic Programming

Name: Genetic Programming | 15th European Conference, EuroGP 2012, Málaga, Spain, April 11-13, 2012, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

15th European Conference, EuroGP 2012, Málaga, Spain, April 11-13, 2012, Proceedings

Alberto Moraglio Sara Silva Krzysztof Krawiec Penousal Machado Carlos Cotta(Herausgeber*in)

Springer (Verlag)

1. Auflage

Erschienen am 23. März 2012

XII, 279 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-29139-5 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

7244
Preface
Organization
Table of Contents
Oral Presentations
Evolving High-Level Imperative Program Trees with Strongly Formed Genetic Programming
Introduction
Background
Strongly Formed Genetic Programming
Initialisation
Mutation
Crossover
Polymorphism
Syntax
Experiments
Factorial
Fibonacci
Even-n-Parity
Results
Example Solution
Conclusions
References
Android Genetic Programming Framework
Introduction
Related Work
AGP: Android Genetic Programming Framework
Motivation
Design Challenges
Common Data Structures
Improvements
Google Reader Application
Application Purpose
Fitness Definition
Results
Context-Aware Localization
Fitness Function Definition
Results
Discussion and Conclusion
References
Genetic Programming for Generalised Helicopter Hovering Control
Introduction
The Dynamic System
The Neuroevolutionary Approach
Application of Genetic Programming
Results
Conclusions and Future Work
References
Cartesian Genetic Programming for Memristive Logic Circuits
Introduction
Background
Memristors
Cartesian Genetic Programming
Genetic Algorithm
Self-Adaptive Mutation
Experimental Setup
Logic Circuits
Results
Robot Control
Environmental Setup
Results
Conclusions
References
A New, Node-Focused Model for Genetic Programming
Introduction
The SNGP Model
Experimentation
Conclusions
References
Medial Crossovers for Genetic Programming
Introduction
Metric-Based Crossover Operators
Partially Medial Crossover
The Experiment
The Puzzle World
Experiment 1: Properties of Search Operators
Experiment 2: Performance in Evolutionary Search
Discussion
Conclusion
References
Improving Face Detection
Introduction
State of the Art
The Framework
Experimental Setup
Classifier Training
Genetic Programming Engine
Assessing Classifier's Performance
Experimental Results
Conclusion and Future Work
References
Grammar Bias and Initialisation in Grammar Based Genetic Programming
Introduction
Grammatical Evolution
Tree-Adjunct Grammatical Evolution
TAGE Derivation Example
Difficulties with Comparing GP Systems
Initialisation and Transformation Bias
Adjunction Bias
Grammar Transformation Bias
Experiments
Results and Discussion
Initialisation
The Effect of PTAGE
New Adjunction Addresses
Conclusions
References
Improving Relevance Measures Using Genetic Programming
Introduction
Relevance Measures
Deficiency in Handling Multi-modal Distributions
Deficiency in Handling Non-orthogonal Multi-variate Relationships
Deficiency in Handling Epistatic Relationships
Using GP for Partitioning the Input Space
Empirical Evaluation
Synthesising Data
GP Settings and Implementation Details
Results
Conclusion
References
An Investigation of Fitness Sharing with Semantic and Syntactic Distance Metrics
Introduction
Methods
Fitness Sharing
Modifying Fitness Sharing
Syntactic Distance
Semantic Distance
Experimental Settings
Results and Discussion
On the Performance
Parameters Analysis
Conclusions and Future Work
References
Evolving Reusable Operation-Based Due-Date Assignment Models for Job Shop Scheduling with Genetic Programming
Introduction
GP for Evolving DDAMs
Representation
Evaluation
Genetic Operators
Fitness Function
Evolution of DDAMs
Experimental Setting
Job Shop Simulation Environment
GP Parameters
Results
Comparison of DDAMs
GP-ADDAM vs. GP-ODDAM
Typical Examples of Evolved DDAMs
Conclusions
References
Evolving Interpolating Models of Net Ecosystem CO2 Exchange Using Grammatical Evolution
Introduction
Evolutionary Approach
Mapping Process
Experimental Setup
Quality of Data and Input Variables
Evolutionary Setup
Measuring Performance
Results and Analysis
Conclusions
References
Multi-Objective Ant Programming for Mining Classification Rules
Introduction
The Multi-Objective Grammar Based Ant Programming (MOGBAP) Algorithm
Environment and Individual Encoding
Heuristic Measures, Transition Rule and Pheromone Maintenance
Muti-Objective Strategy and Niching Procedure
Experimental Set-Up
Experimental Results
Predictive Accuracy Analysis
Comprehensibility Analysis
Conclusions and Future Work
References
Matrix Analysis of Genetic Programming Mutation
Introduction
The Bin Packing Problem
Previous Work
The Matrix Representation
Genetic Programming Parameters
Distance Metrics
Metric 1
Metric 2
Metric 3
Results
Conclusions
References
An Ecological Approach to Measuring Locality in Linear Genotype to Phenotype Maps
Introduction
Related Work
The Mantel Test
Significance Testing on Genotype-Phenotype Maps
Distance Metrics under the Mantel Statistic
Experiment
Discussion
Conclusions
References
Coevolution in Cartesian Genetic Programming
Introduction
Cartesian Genetic Programming
Coevolution of Fitness Predictors in CGP
Population of Candidate Programs
Set of Trainers
Population of Fitness Predictors
Implementation
Results
Benchmark Problems
Experimental Setup
Comparison of Coevolving CGP with Standard CGP
Discussion
Conclusions
References
Evolutionary Design of Message Efficient Secrecy Amplification Protocols
Introduction
Previous Work
Evolution of Amplification Protocols
LGP Tuning and Exploring the Design Space
Experimental Setup
LGP Performance
Discovering New Group-Oriented Protocols
Long-Running Experiments
Performance of Evolved Secrecy Amplification Protocols
Robustness of Discovered Protocols
Multi-criteria Optimization
Weighted Fitness
Optimizing the Number of Messages
Conclusions
References
Automatic Design of Ant Algorithms with Grammatical Evolution
Introduction
Ant Colony Optimization
The Evolutionary Framework
Grammar Definition
Related Work
Experiments and Analysis
Learning the Architectures
Validation of the Evolved Architectures
Comparison with Standard ACO Algorithms
Conclusions
Posters
Random Sampling Technique for Overfitting Control in Genetic Programming
Introduction
State of the Art
Experiments
Datasets
Random Sampling Technique
Parameters and Statistical Tests
Results and Discussion
Results
Discussion
Conclusions
Evolutionary Operator Self-adaptation with Diverse Operators
Introduction
Background
Operator Parameter Adaptation
Evolutionary Algorithms
Methods and Experiments
New Adaptive Mechanisms
Test Problems
Genetic Algorithm Details
TAG3P System Details
Adaptive Mechanisms
Results
The GA Parameter Fitting Problem
TAG3P Symbolic Regression Problems
Operator Application Rates
Discussion
Conclusions
Summary
Assumptions and Limitations
Further Work
References
The Effect of Bloat on the Efficiency of Incremental Evolution of Simulated Snake-Like Robot
Introduction
Sidewinding and Sensing Snake-Like Modular Robot
Evolutionary Framework and the Simulation Environment
Experiments
Stage 1: Evolution of Fast Moving Snakebots from Random Population
Stage 2: Seeded Evolution of Sensing Fast Moving Snakebots
Conclusions
References
Bayesian Network Structure Learning from Limited Datasets through Graph Evolution
Introduction
Background
Bayesian Networks
Akaike Information Criterion
Bayesian Network Structure Learning
Proposed Methodology
GP
Individual Encoding
Fitness Function
Case Study
Experimental Results and Discussion
Conclusions
References
Efficient Phenotype Evaluation in Cartesian Genetic Programming
Introduction
Cartesian Genetic Programming
Fitness Function Evaluation
Common Linear CGP Interpreter
Proposed Method
Data Execution Prevention Handling
Machine Code Generation
Translating Primitive Functions into Machine Code
Translating CGP Phenotype into Machine Code
Machine Code Vectorization
Experimental Setup
Experimental Results and Discussion
Conclusion
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken