Advances in Computer Science and Information Technology. Computer Science and Engineering. Second International Conference, CCSIT 2012, Bangalore, India, January 2-4, 2012. Proceedings, Part II

Research Article

A Non-revisiting Genetic Algorithm with Adaptive Mutation for Function Optimization

Download
437 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-27308-7_31,
        author={Saroj and Devraj},
        title={A Non-revisiting Genetic Algorithm with Adaptive Mutation for Function Optimization},
        proceedings={Advances in Computer Science and Information Technology. Computer Science and Engineering. Second International Conference, CCSIT 2012, Bangalore, India, January 2-4, 2012. Proceedings, Part II},
        proceedings_a={CCSIT PATR II},
        year={2012},
        month={11},
        keywords={Function optimization Genetic algorithm Non-revisiting Adaptive mutation},
        doi={10.1007/978-3-642-27308-7_31}
    }
    
  • Saroj
    Devraj
    Year: 2012
    A Non-revisiting Genetic Algorithm with Adaptive Mutation for Function Optimization
    CCSIT PATR II
    Springer
    DOI: 10.1007/978-3-642-27308-7_31
Saroj1,*, Devraj1,*
  • 1: Guru Jambheshwar University of Science and Technology
*Contact email: ratnoo.saroj@gmail.com, devraj.kamboj@gmail.com

Abstract

Genetic Algorithm (GA) is a robust and popular stochastic optimization algorithm for large and complex search spaces. The major disadvantages of Genetic Algorithms are premature convergence and revisits to individual solutions in the search space. In other words, Genetic algorithm is a revisiting algorithm that leads to duplicate function evaluations which is a clear waste of time and computational resources. In this paper, a non-revisiting genetic algorithm with adaptive mutation is proposed for the domain of function optimization. In this algorithm whenever a revisit occurs, the underlined search point is replaced with a mutated version of the best/random (chosen probabilistically) individual from the GA population. Moreover, the suggested approach is not using any extra memory resources to avoid revisits. To test the power of the method, the proposed non-revisiting algorithm is evaluated using nine benchmarks functions. The performance of the proposed genetic algorithm is superior as compared to simple genetic algorithm as confirmed by the experimental results.