EFFECTIVE AND EFFICIENT CACHING IN GENETIC ALGORITHMS

2001 ◽  
Vol 10 (01n02) ◽  
pp. 273-301 ◽  
Author(s):  
EUNICE E. SANTOS ◽  
EUGENE SANTOS
Keyword(s):  
Genetic Algorithms ◽  
Optimization Problems ◽  
Formal Model ◽  
Solution Space ◽  
Divide And Conquer ◽  
Rigorous Analysis ◽  
Fitness Functions ◽  
Large Numbers ◽  
Randomized Methods ◽  
Discrete Optimization Problems

Hard discrete optimization problems using randomized methods such as genetic algorithms require large numbers of samples from the solution space. Each candidate sample/solution must be evaluated using the target fitness/energy function being optimized. Such fitness computations are a bottleneck in sampling methods such as genetic algorithms. We observe that the caching of partial results from these fitness computations can reduce this bottleneck. We provide a rigorous analysis of the run-times of GAs with and without caching. By representing fitness functions as classic Divide and Conquer algorithms, we provide a formal model to predict the efficiency of caching GAs vs. non-caching GAs. Finally, we explore the domain of protein folding with GAs and demonstrate that caching can significantly reduce expected run-times through both theoretical and extensive empirical analyses.

A fast genetic algorithm for solving architectural design optimization problems

2015 ◽  
Vol 29 (4) ◽  
pp. 457-469 ◽  
Author(s):  
Zhouzhou Su ◽  
Wei Yan
Keyword(s):  
Genetic Algorithms ◽  
Design Optimization ◽  
Domain Knowledge ◽  
Architectural Design ◽  
Optimization Problems ◽  
Computing Time ◽  
Divide And Conquer ◽  
Specific Domain ◽  
Building Performance Simulation ◽  
Simulation And Optimization

AbstractBuilding performance simulation and genetic algorithms are powerful techniques for helping designers make better design decisions in architectural design optimization. However, they are very time consuming and require a significant amount of computing power. More time is needed when two techniques work together. This has become the primary impediment in applying design optimization to real-world projects. This study focuses on reducing the computing time in genetic algorithms when building simulation techniques are involved. In this study, we combine two techniques (offline simulation and divide and conquer) to effectively improve the run time in these architectural design optimization problems, utilizing architecture-specific domain knowledge. The improved methods are evaluated with a case study of a nursing unit design to minimize the nurses’ travel distance and maximize daylighting performance in patient rooms. Results show the computing time can be saved significantly during the simulation and optimization process.

RDF Query Path Optimization Using Hybrid Genetic Algorithms

2022 ◽  
Vol 12 (1) ◽  
pp. 1-16
Author(s):  
Qazi Mudassar Ilyas ◽  
Muneer Ahmad ◽  
Sonia Rauf ◽  
Danish Irfan
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Execution Time ◽  
Optimization Problems ◽  
Hybrid Genetic Algorithm ◽  
Solution Space ◽  
Path Optimization ◽  
Initial Population ◽  
Description Framework ◽  
The Cost

Resource Description Framework (RDF) inherently supports data mergers from various resources into a single federated graph that can become very large even for an application of modest size. This results in severe performance degradation in the execution of RDF queries. As every RDF query essentially traverses a graph to find the output of the Query, an efficient path traversal reduces the execution time of RDF queries. Hence, query path optimization is required to reduce the execution time as well as the cost of a query. Query path optimization is an NP-hard problem that cannot be solved in polynomial time. Genetic algorithms have proven to be very useful in optimization problems. We propose a hybrid genetic algorithm for query path optimization. The proposed algorithm selects an initial population using iterative improvement thus reducing the initial solution space for the genetic algorithm. The proposed algorithm makes significant improvements in the overall performance. We show that the overall number of joins for complex queries is reduced considerably, resulting in reduced cost.

Researches on Function Optimization Problems Based on Chaotic Immune Genetic Algorithms

2012 ◽  
Vol 616-618 ◽  
pp. 2064-2067
Author(s):  
Yong Gang Che ◽  
Chun Yu Xiao ◽  
Chao Hai Kang ◽  
Ying Ying Li ◽  
Li Ying Gong
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Optimization Problems ◽  
Chaotic Map ◽  
Solution Space ◽  
Population Diversity ◽  
Function Optimization ◽  
Initial Population ◽  
Immune Genetic Algorithm ◽  
Complex Functions

To solve the primary problems in genetic algorithms, such as slow convergence speed, poor local searching capability and easy prematurity, the immune mechanism is introduced into the genetic algorithm, and thus population diversity is maintained better, and the phenomena of premature convergence and oscillation are reduced. In order to compensate the defects of immune genetic algorithm, the Hénon chaotic map, which is introduced on the above basis, makes the generated initial population uniformly distributed in the solution space, eventually, the defect of data redundancy is reduced and the quality of evolution is improved. The proposed chaotic immune genetic algorithm is used to optimize the complex functions, and there is an analysis compared with the genetic algorithm and the immune genetic algorithm, the feasibility and effectiveness of the proposed algorithm are proved from the perspective of simulation experiments.

scholarly journals Solving Random Travelling Salesman Problem using Firefly Algorithm

2020 ◽  
Vol 9 (4) ◽  
pp. 1037-1041
Keyword(s):  
Genetic Algorithms ◽  
Simulated Annealing ◽  
Firefly Algorithm ◽  
Ant Colony Algorithm ◽  
Optimization Problems ◽  
Travelling Salesman Problem ◽  
Ant Colony ◽  
Ant Colony Algorithms ◽  
Comparison Of The Results ◽  
Discrete Optimization Problems

The firefly algorithm is a recently developed optimization algorithm, which is suitable for solving any kind of discrete optimization problems. This is an algorithm inspired from the nature. In this paper, a firefly algorithm is proposed to solve random traveling salesman problem. The solution to this problem is already proposed by the algorithms like simulated annealing, genetic algorithms and ant colony algorithms. This algorithm is developed to deal with the issue of accuracy and convergence rate in the solutions provided by those algorithms. A comparison of the results produced by proposed algorithm with the results of simulated annealing, genetic algorithms and ant colony algorithm is given. Finally, the effectiveness of the proposed algorithm is discussed.

scholarly journals Multi-Population Parallel Wolf Pack Algorithm for Task Assignment of UAV Swarm

2021 ◽  
Vol 11 (24) ◽  
pp. 11996
Author(s):  
Yingtong Lu ◽  
Yaofei Ma ◽  
Jiangyun Wang
Keyword(s):  
Population Distribution ◽  
Optimization Problems ◽  
Solution Space ◽  
Task Assignment ◽  
Equal Mass ◽  
Population Diversity ◽  
Simulation Experiments ◽  
Uav Swarm ◽  
Discrete Optimization Problems ◽  
Migration Probability

The effectiveness of the Wolf Pack Algorithm (WPA) in high-dimensional discrete optimization problems has been verified in previous studies; however, it usually takes too long to obtain the best solution. This paper proposes the Multi-Population Parallel Wolf Pack Algorithm (MPPWPA), in which the size of the wolf population is reduced by dividing the population into multiple sub-populations that optimize independently at the same time. Using the approximate average division method, the population is divided into multiple equal mass sub-populations whose better individuals constitute an elite sub-population. Through the elite-mass population distribution, those better individuals are optimized twice by the elite sub-population and mass sub-populations, which can accelerate the convergence. In order to maintain the population diversity, population pretreatment is proposed. The sub-populations migrate according to a constant migration probability and the migration of sub-populations are equivalent to the re-division of the confluent population. Finally, the proposed algorithm is carried out in a synchronous parallel system. Through the simulation experiments on the task assignment of the UAV swarm in three scenarios whose dimensions of solution space are 8, 30 and 150, the MPPWPA is verified as being effective in improving the optimization performance.

Crossover Strategy for Improved Solution Space Exploration With Genetic Algorithms

1998 ◽  
Author(s):  
Kurt S. Anderson ◽  
YuHung Hsu
Keyword(s):  
Genetic Algorithms ◽  
Optimization Problems ◽  
Single Point ◽  
Solution Space ◽  
Nonuniform Distribution ◽  
High Dimensional ◽  
Optimal Solutions ◽  
Crossover Operator ◽  
Entire Chromosome ◽  
Dimensional Optimization

Abstract The following paper presents a modified crossover operator to extend the exploration capability in Genetic Algorithms for high dimensional optimization problems. Traditional strategies apply crossover once on a pair of selected chromosomes to generate two offspring by randomly selecting a single crossover location within the chromosomal length. The proposed method applies crossover once on each separate gene (variable) instead of on the entire chromosome. To further accelerate exploration of the Genetic Algorithm, nonuniform distribution of the respective crossover position on each gene has also been studied. The empirical results show that Genetic Algorithms with the proposed crossover strategies can find optimal or near optimal solutions within fewer generations than traditional single point crossover.

scholarly journals Image Segmentation Using Genetic Algorithms

2021 ◽  
Vol 4 ◽  
pp. 52-55
Author(s):  
Semen Gorokhovskyi ◽  
Andrii Moroz
Keyword(s):  
Image Processing ◽  
Genetic Algorithms ◽  
Image Segmentation ◽  
Optimization Problems ◽  
Random Search ◽  
Solution Space ◽  
Optimization Techniques ◽  
Combinatorial Search ◽  
Good Tool ◽  
Data Set

Image segmentation is a crucial step in the image processing and analysis process. Image segmentation is the process of splitting one image into many segments. Image segmentation divides images into segments that are more representative and easier to examine. Individual surfaces or items can be used as such pieces. The process of image segmentation is used to locate objects and their boundaries.Genetic algorithms are stochastic search methods, the work of which is taken from the genetic laws, natural selection, and evolution of organisms. Their main attractive feature is the ability to solve complex problems of combinatorial search effectively, because the parallel study of solutions, largely eliminates the possibility of staying on the local optimal solution rather than finding a global one.The point of using genetic algorithms is that each pixel is grouped with other pixels using a distance function based on both local and global already calculated segments. Almost every image segmentation algorithm contains parameters that are used to control the segmentation results; the genetic system can dynamically change parameters to achieve the best performance.Similarly to image sequencing, to optimize several parameters in the process, multi-targeted genetic algorithms were used, which enabled finding a diverse collection of solutions with more variables. Multi- targeted Genetic Algorithm (MTGA) is a guided random search method that consists of optimization techniques. It can solve multi-targeted optimization problems and explore different parts of the solution space. As a result, a diversified collection of solutions can be found, with more variables that can be optimized at the same time. In this article several MTGA were used and compared.Genetic algorithms are a good tool for image processing in the absence of a high-quality labeled data set, which is either a result of the long work of many researchers or the contribution of large sums of money to obtain an array of data from external sources.In this article, we will use genetic algorithms to solve the problem of image segmentation.

Basic Study on Assembling of Objective Functions in Many-Objective Optimization Problems

2010 ◽  
Vol 14 (6) ◽  
pp. 618-623
Author(s):  
Shun Otake ◽  
◽  
Tomohiro Yoshikawa ◽  
Takeshi Furuhashi
Keyword(s):  
Optimization Problems ◽  
Solution Space ◽  
Correlation Coefficients ◽  
Scheduling Problem ◽  
Objective Functions ◽  
Nurse Scheduling ◽  
Pareto Solutions ◽  
Basic Study ◽  
Fitness Functions ◽  
Multiobjective Optimization Problems

Genetic Algorithms (GAs) have been widely applied to Multiobjective Optimization Problems (MOPs), called MOGA. A set of Pareto solutions in MOPs having plural fitness functions are searched, then GA is applied in a multipoint search. MOGA performance decreases with the increasing number of objective functions because solution space spreads exponentially. An effective MOGA search is an important issue in many objective optimization problems. One effective approach is assembling objective functions and reducing their number, but appropriate assembly and the number of objective functions to be assembled has not been studied sufficiently. Our purpose here is to determine the effects of assembling objective functions by studying assembly effects when MOGA is applied to a simplified Nurse Scheduling Problem (sNSP) in two types of assembly based on objective function meaning and correlation coefficients.

A Genetic Algorithm for Solving Mixed Discrete Optimization Problems

1993 ◽  
Author(s):  
Zhihang Qian ◽  
Jun Yu ◽  
Ji Zhou
Keyword(s):  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Discrete Optimization ◽  
High Stability ◽  
Optimization Problems ◽  
Global Optimum ◽  
Optimal Method ◽  
Simple Genetic Algorithms ◽  
Discrete Optimization Problems

Abstract A new optimal method based on genetic algorithms (GAs) is proposed here towards the mixed discrete optimization problems. This method has not only the advantages of high stability and wide adaptability but also a better chance of locating the global optimum. Its efficiency is much higher than that of simple genetic algorithms.

Sign in / Sign up

Export Citation Format

Share Document