Categories

Archive

2002 017

Fitness Inheritance in Multi-Objective Optimization

Chen, J.-H., Goldberg, D. E., Hoand, S.-Y., Sastry, K. (2002)
TR No.: 2002017 | Download PDF | Download PS

Abstract:
In real-world multi-objective problems, the evaluation of objective functions usually requires a large amount of computation time. Moreover, due to the curse of dimensionality, solving multiobjective problems often requires much longer computation time than solving single-objective problems. Therefore, it is essential to develop e?ciency enhancement techniques for solving multi-objective problems. This paper investigates fitness inheritance […]

2002 016

Accuracy, Parsimony, and Generality in Evolutionary LearningSystems via Multiobjective Selection

Llorà, X., Goldberg, D. E., Traus, I., Bernadó, E. (2002)
TR No.: 2002016 | Download PDF | Download PS

Abstract:
Learning systems (also known as Pittsburgh learning classifier systems) need to balance accuracy and parsimony for evolving high quality general hypotheses. The evolutionary learning process used in learning systems is based on using a set of training instances that sample the target concept to be learned. Thus, the the learning process may overfit the learned […]

2002 015

Minimal Achievable Error in the LED problem

Llorà, X., Goldberg, D. E. (2002)
TR No.: 2002015 | Download PDF | Download PS

Abstract:
This paper presents a theoretical model to predict the minimal achievable error, given a noise ratio, in the LED data set problem. The motivation for developing this theoretical model is to understand and explain some of the results that different systems achieve when they solve the LED problem. Moreover, given a new learning algorithm that […]

2002 014

The Influence of Binary Representations of Integers on thePerformance of Selectorecombinative Genetic Algorithms

Rothlauf, F. (2002)
TR No.: 2002014 | Download PDF | Download PS

Abstract:
When using representations for genetic algorithms (GAs) every optimization problem can be separated into a genotype-phenotype and a phenotype-fitness mapping. The genotype-phenotype mapping is the used representation and the phenotype-fitness mapping is the problem that should be solved.
This paper investigates how the use of different binary representations of integers influences the performance of selectorecombinative GAs […]

2002 011

Theory of Generalization and Learning in XCS

Butz, M., Kovacs, T., Lanzi, P. L., Wilson, S. W. (2002)
TR No.: 2002011 | Download PDF | Download PS

Abstract:
The XCS classifier system evolves accurate, maximally general solutions to a wide variety of machine learning, data, and robotics problems, but a theoretical basis for these properties has not been presented. This paper develops and tests a model for the generalization pressure, and establishes conditions for the existence of effective fitness or accuracy pressure in […]

2002 013

How Well Does a Single-Point Crossover Mixing Building Blocks with Tight Linkage?

Sastry, K., Goldberg, D. E. (2002)
TR No.: 2002013 | Download PDF | Download PS

Abstract:
Ensuring building-block (BB) mixing is critical to the success of genetic and evolutionary algorithms. This study develops facetwise models to predict the BB mixing time and the population sizing dictated by BB mixing for single-point crossover. Empirical results are used to validate these models. The population-sizing model suggests that for moderate-to-large problems, BB mixing—instead of […]

2002 012

Analysis of Mixing in Genetic Algorithms: A Survey

Sastry, K., Goldberg, D. E. (2002)
TR No.: 2002012 | Download PDF | Download PS

Abstract:
Ensuring building-block (BB) mixing is critical to the success of genetic and evolutionary algorithms. There has been a growing interest in analyzing and understanding BB mixing and it is necessary to organize and categorize representative literature. This paper presents an exhaustive survey of studies on one or more aspects of mixing. In doing so, a […]

2002 010

Genetic Algorithms at the University of Illinois Fall 2001

Goldberg, D. E. (ed.) (2002)
TR No.: 2002010 | Download PDF | Download PS

Abstract:
Related PostsGenetic Algorithms at the University of Illinois Fall 1999Genetic Algorithms at the University of Illinois Fall 2000Genetic Algorithms at the University of Illinois Fall 2002

2002 009

Multi-objective Bayesian Optimization Algorithm

Khan, N., Goldberg, D.E., Pelikan, M. (2002)
TR No.: 2002009 | Download PDF | Download PS

Abstract:
This paper proposes a competent multi-objective genetic algorithm
called the multi-objective Bayesian optimization algorithm
(mBOA). mBOA incorporates the selection method of the non-dominated
sorting genetic algorithm-II (NSGA-II) into the Bayesian optimization
algorithm (BOA). The proposed algorithm has been tested on an array of
test functions which incorporate deception and loose-linkage and the
results are compared to those of NSGA-II. Results indicate […]

2002 008

XCS with Average Reward Criterion in Multi-Step Environment.

Tharakunnel, K., Goldberg, D.E. (2002)
TR No.: 2002008 | Download PDF | Download PS

Abstract:
In multi-step environment, the XCS prediction parameter is an estimate of the discounted sum of
successive rewards. Thus, XCS isdesigned to address sequential (multi-step) decision problems
with the objective of maximization of the total discounted rewards. However, there are many
real-world sequential decision problems where the preferred objective is maximization of the average of successive rewards. In […]