Genetic Programming: 4th European Conference, EuroGP 2001 Lake Como, Italy, April 18–20, 2001 Proceedings (Lecture Notes in Computer Science, 2038)

Genetic Programming
Preface
Organisation
Table of Contents
Heuristic Learning Based on Genetic Programming
	Introduction
	Problem Description
		BOMs
	GP Solution
		Representation
		Operators
		Algorithm and Parameter Setting
		Size Reduction
	Experimental Results
	Conclusions
	References
Evolving Color Constancy for an Artificial Retina
	Introduction
	An Artificial Retina
	Experiments
	Results
	Conclusion
	References
Adaptive Genetic Programming Applied to New and Existing Simple Regression Problems
	Introduction
	Symbolic Regression with Genetic Programming
	Stepwise Adaptation of Weights
	Problem Generator
	Experiments and Results
		Koza Functions
		Randomly Generated Polynomials
	Conclusions
	Future Research
	References
An Evolutionary Approach to Automatic Generation of VHDL Code for Low-Power Digital Filters
	Introduction
	Motivation and Problem Statement
	Genetic Representation of FIR Filters
	The Evolutionary Algorithm
	Experiments and Results
		Low-Pass Filter with Normalized Pass Band (0, 0.25)
		Low-Pass Filter with Normalized Pass Band (0, 0.35)
	Conclusion
	References
Studying the Influence of Communication Topology and Migration on Distributed Genetic Programming
	Introduction
	Parallel and Distributed Genetic Programming (PADGP)
		Implementation and Software Tools
	Methodology and Problems Description
		Effort of Computation
		Even Parity 5 Problem
		The Artificial Ant Problem on the Santa Fe Trail
		The Field Programmable Gate Array (FPGA) Problem
	Comparing Topologies
	Studying Migration Parameters
	Conclusions
	References
CAGE: A Tool for Parallel Genetic Programming Applications
	Introduction
	Parallel Genetic Programming
	Parallel Implementation of CAGE
	Experimental Results
	Conclusions and Future Work
	References
Ripple Crossover in Genetic Programming
	Introduction
	Context Free Grammars and Grammatical Evolution
	Closed vs. Context Free Grammars
	Ripple Crossover in GE and GP
	Experiments
	Results
	Discussion
	Conclusions
	References
Evolving Receiver Operating Characteristics for Data Fusion
	Introduction
	``Maximum Realisable\'\' ROC
	An Example
	Evolving a Combined Classifier
		Function and Terminal Sets
		Fitness Function
		Results
	Discussion
	Conclusions
	Future Work
	References
An Adaptive Mapping for Developmental Genetic Programming
	Introduction
	Stack-Based Genetic Programming
	An Adaptive Genotype to Phenotype Mapping
		Using a Huffman-Decoding Mapping
		A Coevolutionary Model for Adaptive Mappings
	Experimental Setup
	Results
	Conclusions
	Future Work
	References
A Schema Theory Analysis of the Evolution of Size in Genetic Programming with Linear Representations
	Introduction
	Schema Theory for GP on Linear Structures
		Three Theories of Bloat
		Why Theory Matters
		Schema Theory Definitions
		The Schema Theorem
		An Exact Theory of Bloat
	Flat Fitness Landscapes
	The One-then-Zeros Problem
		One-then-Zeros Problem Definition
		Calculating alpha_xo ((=)^N , t) and alpha_xo(1(0)^alpha ,t)
		One-then-Zeros Results
	Landscape Levels and Program Size
		Computing the Mean Size
	Conclusions and Future Work
	References
Exact Schema Theorems for GP with One-Point and Standard Crossover Operating on Linear Structures and Their Application to the Study of the Evolution of Size
	Introduction
	Exact Schema Theory for Linear Structures
	Evolution of Size in Linear Systems
	Experimental Results
	Fixed-Point Size Distribution for Standard Crossover on a Flat Landscape
	Search Space Sampling under Standard Crossover
	Conclusions
	References
General Schema Theory for Genetic Programming with Subtree-Swapping Crossover
	Introduction
	Background
	Node Reference Systems
	Functions over Node Reference Systems
	Modelling Subtree-Swapping Crossovers
	Microscopic Exact GP Schema Theorem for Subtree-Swapping Crossovers
	Macroscopic Exact GP Schema Theorem
	Applications and Specialisations
		Macroscopic Exact Schema Theorem for GP with Standard Crossover
		Size-Evolution Equation for GP with Subtree-Swapping Crossover
		Effective Fitness for GP with Subtree-Swapping Crossovers
	Conclusions
	References
Evolving Modules in Genetic Programming by Subtree Encapsulation
	Introduction
	Subtree Encapsulation Scheme
		Selecting Subtrees for Modularisation
		Subtree Database
		Encapsulation Procedure
	Subtree Encapsulation for Target Detection
		Training Data
		GP Specifications
		Target Detection Performance
		Varying the Number of Encapsulated Subtrees
		Computation Time
		Interpretation of Subtree Operation
	Conclusions
	References
Evolution of Affine Transformations and Iterated Function Systems Using Hierarchical Evolution Strategy
	Introduction
	Hierarchical Evolution Strategy
		Strongly Typed Genetic Programming
	Evolution of Affine Transformations
		STGP Architecture
		Architecture Using the Hierarchical Evolution Strategy
		Mutation for ES Individuals, Fitness Function, and Control Parameters
		Results for the Inverse Problem for Affine Transformations
	Evolution of Iterated Function Systems
		Architecture Using the Hierarchical Evolution Strategy
		Architecture Using STGP
		Fitness Function
		Control Parameters
		Results for the Inverse Problem for IFS
	Conclusions
	References
Evolving Turing Machines for Biosequence Recognition and Analysis
	Introduction
	Turing Machines as the Model of Computation of GP
	Experiments
		Experiment 1: Turing Machines
		Experiment 2: Two-Way Deterministic Finite Automata
		Experiment 3: Multiple Sequence Alignment
	Conclusions and Future Work
	References
Neutrality and the Evolvability of Boolean Function Landscape
	Introduction
	Neutrality: Implicit Versus Explicit
		Neutral Versus Adaptive Mutation
	Cartesian Genetic Programming
	Experiments
		Even-3-Parity Problem
		Evolutionary Algorithm
		Neutrality Measured with Hamming Distance
	Results
	Analysis and Discussion
	Related Work
	Conclusion and Future Work
	Referernces
Programmable Smart Membranes: Using Genetic Programming to Evolve Scalable Distributed Controllers for a Novel Self-Reconfigurable Modular Robotic Application
	Introduction
	Related Work
	Modular Robot Simulator
		Movement
		Sensors
	The Smart Membrane Problem
		Function Set
		Terminal Set
		Fitness
		Parallelization
	Results
	Generalization
	Conclusions and Future Work
	References
A GP Artificial Ant for Image Processing: Preliminary Experiments with EASEA
	Introduction
		EASEA and genetic programming
		The artificial ant problem, animats and low-level image processing
	Implementing a food foraging process using EASEA
		Animat functions
		Initialising trees
		Crossover
		Mutation
		Fitness evaluation
		Results
	Contour detection
		Grey level images and contour detection
		Basic animat functions
		Genetic operators
		Results
	Comments on implementation
	Conclusion and future work
	References
Feature Extraction for the k-Nearest Neighbour Classifier with Genetic Programming
	Introduction
	Related Work
	A Framework for Feature Extraction
		Overview
		Features
		Course of Evolution
		Fitness Measure
		Stopping Criterion
		Comparison Algorithms
	Experiments
	Results
		Evolved Features
		Timing
	Conclusion
	References
An Indirect Block-Oriented Representation for Genetic Programming
	Introduction
	GP Codings for Models or Controller
		Direct Block-Oriented Representation
		Indirect Block-Oriented Representation
	Comparison of Both Representations
	Real World Application of the Indirect Representation
	Conclusion
	References
Raising the Dead: Extending Evolutionary Algorithms with a Case-Based Memory
	Introduction
	Context
	Periodically Changing Problems
	Incorporating a Case-Based Memory
	Experiments
	Conclusion and Future Work
	References
Layered Learning in Genetic Programming for a Cooperative Robot Soccer Problem
	Introduction
	Keep-Away Soccer
	Layered Learning
	Experiments
	Results
	Conclusions
	References
Linear-Tree GP and Its Comparison with Other GP Structures
	Introduction of Linear-Tree GP
		Recombination of Linear-Tree Programs
		Mutation
	Test Problems
		Symbolic Regression
		Non-regression Problems
	Experimental Results
		Difference between the Representations
		Analysis of the Linear-Tree-Structure
		The Analysis of the Crossover-Operator
	Summary and Outlook
	References
Evolving Hand-Eye Coordination for a Humanoid Robot with Machine Code Genetic Programming
	Introduction
	Background
	The Elvis Humanoid
		Software Architecture
		Experiments
	Inverse Kinematics Experiment
		Cameras and Image Processing
		Collection of Training Data
		Use of Discipulus
		Demonstration
		Results
	Discussion
	Future Work
	Summary and Conclusions
	References
Adaption of Operator Probabilities in Genetic Programming
	Introduction
	GGP and Genetic Operators
		Graph Representation
		Genetic Operators
		The Evolutionary Algorithm
	Different Methods of Adaption
		Population-Level Dynamic Probabilities (PDP)
		Fitness Based Dynamic Probabilities (FBDP)
		Individual-Level Dynamic Probabilities (IDP)
	Experiments and Results
		Symbolic Regression
		Classification
	Discussion
	Conclusion
	References
Crossover in Grammatical Evolution: The Search Continues
	Introduction
	Grammatical Evolution
	The GE Crossover Operator
	Experimental Approach
	Results
	Discussion
	Conclusions and Future Work
	References
	Symbolic Regression Grammar
	Santa Fe Trail Grammar
Computational Complexity, Genetic Programming, and Implications
	Introduction
	Kolmogorov Complexity
	GP Complexity
		GPs for Programs
		GPs for Function Optimization
	Implications to GP Design
		Large vs. Small Populations
		Increasing Complexity
		Quantum GPs
		Universal Genetic Program
	Conclusions
	References
Genetic Programming for Financial Time Series Prediction
	Introduction
	The Problem
	The Algorithm
		The Initial Structures
		Fitness
		Operations for Modifying Structures
	Experiments
		Test Problem: Parabola
		Predicting the Dow Jones
	Conclusion
	References
Active Handwritten Character Recognition Using Genetic Programming
	Introduction
	Related Work
	GeneARM- Genetic Active Recognition Methods
		Apparatus Setup
		Fitness Testing Phase
	Experiments
	Results and Discussion
	Conclusion
	References
Author Index
Subject Index
3540453555_17_OnlinePDF.pdf.pdf
	1 Introduction
	2 Types in Genetic Programming
	3 Dynamically Typed GP
	4 Strongly Typed GP
	5 Experiments
	6 Discussion
	7 Conclusions
	Bibliography