TY - CONF

T1 - Self-adaptive hybrid genetic algorithm using an ant-based algorithm

AU - El-Mihoub, Tarek

AU - Hopgood, Adrian

AU - Aref, Ibrahim

PY - 2015/10/12

Y1 - 2015/10/12

N2 - The pheromone trail metaphor is a simple and effective way to accumulate the experience of the past solutions in solving discrete optimization problems. Ant-based optimization algorithms have been successfully employed to solve hard optimization problems. The problem of achieving an optimal utilization of a hybrid genetic algorithm search time is actually a problem of finding its optimal set of control parameters. In this paper, a novel form of hybridization between an ant-based algorithm and a genetic-local hybrid algorithm is proposed. An ant colony optimization algorithm is used to monitor the behavior of a genetic-local hybrid algorithm and dynamically adjust its control parameters to optimize the exploitation-exploration balance according to the fitness landscape.

AB - The pheromone trail metaphor is a simple and effective way to accumulate the experience of the past solutions in solving discrete optimization problems. Ant-based optimization algorithms have been successfully employed to solve hard optimization problems. The problem of achieving an optimal utilization of a hybrid genetic algorithm search time is actually a problem of finding its optimal set of control parameters. In this paper, a novel form of hybridization between an ant-based algorithm and a genetic-local hybrid algorithm is proposed. An ant colony optimization algorithm is used to monitor the behavior of a genetic-local hybrid algorithm and dynamically adjust its control parameters to optimize the exploitation-exploration balance according to the fitness landscape.

U2 - 10.1109/ROMA.2014.7295881

DO - 10.1109/ROMA.2014.7295881

M3 - Conference paper

ER -