Genetic algorithm behavior in the MAXSAT domain

Rana, Soraya B.; Whitley, L. Darrell

doi:10.1007/bfb0056920

Cited by 21 publications

(16 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The goal is to find a setting for these variables, i.e., an element of the solution space X = {0, 1} v so that B becomes true. According to [35], MAX-SAT problems are deceptive with many neutral regions. For a Genetic Algorithm, it is therefore hard to find the global optimum.…”

Section: B Combinatorial Optimization: Max-3satmentioning

confidence: 99%

See 1 more Smart Citation

Frequency Fitness Assignment

Weise

Wan

Wang

et al. 2014

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Abstract-Metaheuristic optimization procedures such as Evolutionary Algorithms are usually driven by an objective function which rates the quality of a candidate solution. However, it is not clear in practice whether an objective function adequately rewards intermediate solutions on the path to the global optimum and it may exhibit deceptiveness, epistasis, neutrality, ruggedness, and a lack of causality. In this paper, we introduce the Frequency Fitness H, subject to minimization, that rates how often solutions with the same objective value have been discovered so far. The ideas behind this method are that good solutions are hard to find and that if an algorithm gets stuck at a local optimum, the frequency of the objective values of the surrounding solutions will increase over time, which will eventually allow it to leave that region again. We substitute a Frequency Fitness Assignment process (FFA) for the objective function into several different optimization algorithms. We conduct a comprehensive set of experiments: the synthesis of algorithms with Genetic Programming (GP), the solution of MAX-3SAT problems with Genetic Algorithms, classification with Memetic Genetic Programming, and numerical optimization with a (1+1) Evolution Strategy, in order to verify the utility of FFA. Given that they have no access to the original objective function at all, it is surprising that for some problems (e.g., the algorithm synthesis task) the FFAbased algorithm variants perform significantly better. However, this cannot be guaranteed for all tested problems. We thus also analyze scenarios where algorithms using FFA do not perform better than with the original objective functions.

show abstract

Section: B Combinatorial Optimization: Max-3satmentioning

confidence: 99%

“…From [35], we know that GAs are weak in solving these problems. The best six GA-DIR setups (all with a population size n = 50 000) can solve at most 5 of 100 cases, which is more likely the result of random mutations than targeted search.…”

Section: ) Experimental Settingsmentioning

confidence: 99%

Frequency Fitness Assignment

Weise

Wan

Wang

et al. 2014

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

show abstract

“…Some of the algorithms in the literature are SAWEA [13] [8], RFEA [14][15], FlipGa [16] and ASAP [17], although a genetic algorithm can be used too [5].…”

Section: Related Workmentioning

confidence: 99%

“…Thus, the goal is not to find a solution but to find the solution that the more constraints satisfies (the more number of matching edges). These kind of problems are know as MAXSAT [4] and here is where an evolutionary algorithm can be useful [5].…”

Section: Introductionmentioning

confidence: 99%

Evolutionary techniques in a constraint satisfaction problem: Puzzle Eternity II

Muñoz

Gutiérrez

Sanchís

2009

2009 IEEE Congress on Evolutionary Computation

View full text Add to dashboard Cite

Abstract-This work evaluates three evolutionary algorithms in a constraint satisfaction problem. Specifically, the problem is the Eternity II, a edge-matching puzzle with 256 unique square tiles that have to be placed on a square board of 16 × 16 cells. The aim is not to completely solve the problem but satisfy as many constraints as possible. The three evolutionary algorithms are: genetic algorithm, an own implementation of a technique based on immune system concepts and a multiobjective evolutionary algorithm developed from the genetic algorithm. In addition to comparing the results obtained by applying these evolutionary algorithms, they also will be compared with an exhaustive search algorithm (backtracking) and random search. For the evolutionary algorithms two different fitness functions will be used, the first one as the score of the puzzle and the second one as a combination of the multiobjective algorithm objectives. We also used two ways to create the initial population, one randomly and the other with some domain information.

show abstract

“…This result was confirmed experimentally in [9], they reported scarce performance of classical GAs when compared with local search methods. In addition, Rana and Whitley showed pure GAs being unsuitable for the MAXSAT fitness function, which counts the number of satisfied clauses, because the corresponding domain contains misleading low-order schema information, and the search space tends to result in similar schema fitness averages [26]. Recent results showed that GAs can nevertheless yield good results for SAT if equipped with additional techniques.…”

Section: Gas and The Satisfiability Problemmentioning

confidence: 99%

Improving the Performance of a Genetic Algorithm Using a Variable-Reordering Algorithm

Rodríguez-Tello

Torres-Jiménez

2004

Genetic and Evolutionary Computation – GECCO 2004

View full text Add to dashboard Cite

Abstract. Genetic algorithms have been successfully applied to many difficult problems but there have been some disappointing results as well. In these cases the choice of the internal representation and genetic operators greatly conditions the result. In this paper a GA and a reordering algorithm were used for solve SAT instances. The reordering algorithm produces a more suitable encoding for a GA that enables a GA performance improvement. The attained improvement relies on the building-block hypothesis, which states that a GA works well when short, low-order, highly-fit schemata (building blocks) recombine to form even more highly fit higher-order schemata. The reordering algorithm delivers a representation which has the most related bits (i.e. Boolean variables) in closer positions inside the chromosome. The results of experimentation demonstrated that the proposed approach improves the performance of a simple GA in all the tests accomplished. These experiments also allow us to observe the relation among the internal representation, the genetic operators and the performance of a GA.

show abstract

Genetic algorithm behavior in the MAXSAT domain

Cited by 21 publications

References 4 publications

Frequency Fitness Assignment

Frequency Fitness Assignment

Evolutionary techniques in a constraint satisfaction problem: Puzzle Eternity II

Improving the Performance of a Genetic Algorithm Using a Variable-Reordering Algorithm

Contact Info

Product

Resources

About