Results 1 -
9 of
9
Spurious Dependencies and EDA Scalability
, 2010
"... Numerous studies have shown that advanced estimation of distribution algorithms (EDAs) often discover spurious (unnecessary) dependencies. Nonetheless, only little prior work exists that would study the effects of spurious dependencies on EDA performance. This paper examines the effects of spurious ..."
Abstract
-
Cited by 5 (2 self)
- Add to MetaCart
(Show Context)
Numerous studies have shown that advanced estimation of distribution algorithms (EDAs) often discover spurious (unnecessary) dependencies. Nonetheless, only little prior work exists that would study the effects of spurious dependencies on EDA performance. This paper examines the effects of spurious dependencies on the performance and scalability of EDAs with the main focus on EDAs with marginal product models and the onemax problem. A theoretical model is proposed to analyze the effects of spurious dependencies on the population sizing in EDAs and the theory is verified with experiments. The effects of spurious dependencies on the number of generations are studied empirically.
General
"... This paper investigates the adaptability of genotypes produced by a modified genome growth algorithm (MGG) in a genetic algorithm. MGG uses constructional selection, a genome growth strategy modeled on the evolution of the biological genome [1]. Experiments did not favor the more costly MGG genotype ..."
Abstract
- Add to MetaCart
(Show Context)
This paper investigates the adaptability of genotypes produced by a modified genome growth algorithm (MGG) in a genetic algorithm. MGG uses constructional selection, a genome growth strategy modeled on the evolution of the biological genome [1]. Experiments did not favor the more costly MGG genotypes. This paper also proposes a new type of mutation operator: meth, which is loosely based on DNA methylation. Experiments suggest the utility of ‘methylation ’ to improve the adaptability of representations and to build robust and adaptable digital structures.
Network EDAs: An Empirical Study
, 2012
"... Learning a good model structure is important to the efficient solving of problems by estimation of distribution algorithms. In this paper we present the results of a series of experiments, applying a structure learning algorithm for undirected probabilistic graphical models based on statistical depe ..."
Abstract
- Add to MetaCart
(Show Context)
Learning a good model structure is important to the efficient solving of problems by estimation of distribution algorithms. In this paper we present the results of a series of experiments, applying a structure learning algorithm for undirected probabilistic graphical models based on statistical dependency tests to three fitness functions with different selection operators, proportions and pressures. The number of spurious interactions found by the algorithm are measured and reported. Truncation selection, and its complement (selecting only low fitness solutions) prove quite robust, resulting in a similar number of spurious dependencies regardless of selection pressure. In contrast, tournament and fitness proportionate selection are strongly affected by the selection proportion and pressure.
Genetic Algorithms and . . . MODELING: APPLICATIONS IN MATERIALS SCIENCE AND CHEMISTRY AND ADVANCES IN SCALABILITY
, 2007
"... Effective and efficient multiscale modeling is essential to advance both the science and synthesis in a wide array of fields such as physics, chemistry, materials science, biology, biotechnology and pharmacology. This study investigates the efficacy and potential of using genetic algorithms for mult ..."
Abstract
- Add to MetaCart
Effective and efficient multiscale modeling is essential to advance both the science and synthesis in a wide array of fields such as physics, chemistry, materials science, biology, biotechnology and pharmacology. This study investigates the efficacy and potential of using genetic algorithms for multiscale materials modeling and addresses some of the challenges involved in designing competent algorithms that solve hard problems quickly, reliably and accurately. In particular, this thesis demonstrates the use of genetic algorithms (GAs) and genetic programming (GP) in multiscale modeling with the help of two non-trivial case studies in materials science and chemistry. The first case study explores the utility of genetic programming (GP) in multi-timescaling alloy kinetics simulations. In essence, GP is used to bridge molecular dynamics and kinetic Monte Carlo methods to span orders-of-magnitude in simulation time. Specifically, GP is used to regress symbolically an inline barrier function from a limited set of molecular dynamics simulations to enable kinetic Monte Carlo that simulate seconds of real time. Results on a non-trivial example of vacancy-assisted migration on a surface of a face-centered cubic (fcc) Copper-Cobalt (CuxCo1−x) alloy show that GP predicts all barriers with 0.1 % error from calculations for less than 3 % of active
Acknowledgement Advisor
, 2011
"... ii Existing estimation of distribution algorithms (EDAs) learn linkages starting from pairwise interactions of variables and construct models from the linkages. The char-acteristic function of models which indicates the relations among variables are bi-nary. In other words, the characteristic functi ..."
Abstract
- Add to MetaCart
(Show Context)
ii Existing estimation of distribution algorithms (EDAs) learn linkages starting from pairwise interactions of variables and construct models from the linkages. The char-acteristic function of models which indicates the relations among variables are bi-nary. In other words, the characteristic function indicates that there exist or not interactions among variables. Empirically, it can occur that two variables should be sometimes related but sometimes not. This thesis introduces a real-valued charac-teristic function to illustrate this property of fuzziness. We examine all the possible binary models and real-valued models on test problems. The results show that EDAs using optimal real-valued models outperforms the one using optimal binary models. This thesis also proposes two recombination algorithms which are able to utilize the information provided by real-valued models. Experiments show that the proposed pairwise crossover could reduce function evaluations by three quarters. Moreover, this thesis proposes an effective method to find a threshold for entropy based linkage-learning metric and a method to generate real-valued models. Experiments show that the proposed crossover with generated real-valued models works well.
Design of Test Problem for Genetic Algorithm
, 2013
"... National Taiwan University When the estimation of distribution algorithms are applied to real-world problems, two kinds of problem structures, overlapping and conflict structures, might be difficult to solve. To test different EDAs ’ capabilities of dealing with overlapping and conflict structures, ..."
Abstract
- Add to MetaCart
(Show Context)
National Taiwan University When the estimation of distribution algorithms are applied to real-world problems, two kinds of problem structures, overlapping and conflict structures, might be difficult to solve. To test different EDAs ’ capabilities of dealing with overlapping and conflict structures, some test problems have been proposed. However, the upper-bound of the degree of overlap and the effect of conflict have not been fully investigated. This paper investigates how to properly define the degree of overlap and the degree of conflict to reflect the difficulties of problems for the EDAs. A new test problem is proposed with the new definitions of the degree of overlap and the degree of conflict. A framework for building the proposed problem is presented, and some model-building genetic algorithms are tested by the problem. This test problem can be applied to further researches on overlapping and conflict structures. 1
The essence of Real-valued Characteristic Function for Pairwise Relation in Linkage Learning for EDAs
, 2011
"... Existing EDAs learn linkages starting from pairwise interactions. The characteristic function which indicates the relations among variables is binary. In other words, the characteristic function indicates that there exist or no interactions among variables. Empirically, it can occur that two variabl ..."
Abstract
- Add to MetaCart
(Show Context)
Existing EDAs learn linkages starting from pairwise interactions. The characteristic function which indicates the relations among variables is binary. In other words, the characteristic function indicates that there exist or no interactions among variables. Empirically, it can occur that two variables should be sometimes related but sometimes not. This paper introduces a real-valued characteristic function to illustrate this property of fuzziness. We examine all the possible binary models and real-valued models on a test function. The results show that the optimal real-valued model is better than all the binary models. This paper also proposes a crossover method which is able to utilize the real-valued information. Experiments show that the proposed crossover could reduce the number of function evaluations up to four times on test problems. Moreover, this paper proposes an effective method to find a threshold for entropy based interaction-detection metric, and the found threshold can be utilized to provide real-valued models. Experiments show that the proposed crossover with the threshold-finding method works well. 1
A Niching Scheme for EDAs to Reduce Spurious Dependencies
, 2013
"... This paper proposes a niching scheme, the dependency structure matrix restricted tourna-ment replacement (DSMRTR). The restricted tournament replacement (RTR) is a well-known niching scheme in the field of estimation of distribution algorithms (EDAs). However, RTR induces spurious dependencies among ..."
Abstract
- Add to MetaCart
(Show Context)
This paper proposes a niching scheme, the dependency structure matrix restricted tourna-ment replacement (DSMRTR). The restricted tournament replacement (RTR) is a well-known niching scheme in the field of estimation of distribution algorithms (EDAs). However, RTR induces spurious dependencies among variables, which impair the performance of EDAs. This paper utilizes building-block-wise distances to define a new distance metric, the one-niche dis-tance. For those EDAs which provide explicit linkage information, the one-niche distances can be directly incorporated into RTR. For EDAs without such information, DSMRTR constructs a dependency structure matrix via the differential mutual complement to estimate the one-niche distances. Empirical results show that DSMRTR induces fewer spurious dependencies than RTR does while maintaining enough diversity for EDAs.
