Genetic Algorithms in Search, Optimization and Machine Learning
Genetic Algorithms in Search, Optimization and Machine Learning
Computers and Intractability: A Guide to the Theory of NP-Completeness
Computers and Intractability: A Guide to the Theory of NP-Completeness
Selectively Destructive Re-start
Proceedings of the 6th International Conference on Genetic Algorithms
Genetic Algorithm Behavior in the MAXSAT Domain
PPSN V Proceedings of the 5th International Conference on Parallel Problem Solving from Nature
Adaptive Fitness Functions for the Satisfiability Problem
PPSN VI Proceedings of the 6th International Conference on Parallel Problem Solving from Nature
Fitness Landscapes and Evolutionary Algorithms
AE '99 Selected Papers from the 4th European Conference on Artificial Evolution
A Superior Evolutionary Algorithm for 3-SAT
EP '98 Proceedings of the 7th International Conference on Evolutionary Programming VII
Backbone fragility and the local search cost peak
Journal of Artificial Intelligence Research
Evolutionary computation and its applications in neural and fuzzy systems
Applied Computational Intelligence and Soft Computing
| Hi-index | 0.00 |
A stagnation of evolutionary search is frequently associated with missing population diversity. The resulting degradation of the over-all performance can be avoided by applying methods for diversity management. This paper introduces a conceptually simple approach to maintain diversity called partial restart. The basic idea is to re-initialize parts of the population after certain time intervals, thereby raising the probability of escaping from local optima that have dominated the recent search progress. The usefulness of the proposed technique is evaluated empirically in two characteristic problem domains, represented by the satisfiability problem and the onemax problem. The main goal is to identify problem structures where partial restarts are promising, and to gain a better understanding of the relations between different variants of partial restarts.