Large Population Size IGAs with Individuals' Fitness Not Assigned by User

Authors:
Jie Yuan;Dunwei Gong
Affiliations:
School of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou, China 221008;School of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou, China 221008
Venue:
ICIC '08 Proceedings of the 4th international conference on Intelligent Computing: Advanced Intelligent Computing Theories and Applications - with Aspects of Artificial Intelligence
Year:
2008

Citing 3
Cited 2

An introduction to genetic algorithms

An introduction to genetic algorithms
Eye-tracking evolutionary algorithm to minimize user fatigue in IEC applied to interactive one-max problem

Proceedings of the 9th annual conference companion on Genetic and evolutionary computation
Interactive Evolutionary Computation-Based Hearing Aid Fitting

IEEE Transactions on Evolutionary Computation

Impact of individuals' fitness expressions on interactive genetic algorithms' performances

CCDC'09 Proceedings of the 21st annual international conference on Chinese control and decision conference
The problem with evolutionary art is ...

EvoCOMNET'10 Proceedings of the 2010 international conference on Applications of Evolutionary Computation - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

User fatigue problem in traditional interactive genetic algorithms restricts its population size. It is necessary to maintain a large population size to optimize complicated problems. We present a large population size interactive genetic algorithm with individuals' fitness not assigned by user in this paper. The algorithm divides the population into several clusters, and the maximum number of clusters is changeable with the evolution and distribution of the population. The user only evaluates a center individual in each cluster and others' fitness is estimated based on these ones. In addition, to assign a center individual's fitness, we record the time when the user evaluates it satisfactory or unsatisfactory according to his/her sensitiveness, and its fitness is automatically calculated based on the time. Finally, we apply the proposed algorithm to the one-max problem, and compare it with traditional interactive genetic algorithms. The experimental results validate its efficiency.