Sun Xiao-yan

Grid-based knowledge-guided interactive genetic algorithm and its application to curtain design

User fatigue is the main bottleneck of interactive genetic algorithm, influencing its performance in searching and limiting its applications in complicated optimization problems. One of the efficient methodologies is to speed up the algorithms convergence to satisfactory solutions by sufficiently using evolutionary knowledge. A grid-based knowledge-guided interactive genetic algorithm is proposed in this paper so as to alleviate user fatigue with less memory cost and higher computational efficiency. From the view of gene sense unit, two 3-dimensional irregular memory grids are built to store all evolutionary information, including the emerged individuals, their emerged frequency and fitness. Then, the emerged frequency and fitness of each gene sense unit are statistical computed along with the evolution. According to the obtained knowledge of a gene sense unit, the time that the users preference is clear is determined and strategies for using such information to mutate and generate child population are designed. The proposed algorithm is applied to a curtain design system, and the results show its feasibility and efficiency in alleviating user fatigue.

Representative selection for cooperative co-evolutionary genetic algorithms

The performance of cooperative co-evolutionary genetic algorithms is highly affected by the representative selection strategy. But rational method is absent now. Oriented to the shortage, the representative selection strategy is studied based on the parallel implementation of cooperative co-evolutionary genetic algorithms in LAN. Firstly, the active cooperation ideology for representative selection and the dynamical determinate method on cooperation pool size are put forward. The methods for determining cooperation pool size, selecting cooperators and permuting cooperations are presented based on the evolutionary ability of sub-population and distributive performance of the individuals. Thirdly, the implementation steps are given. Lastly, the results of benchmark functions optimization show the validation of the method.The performance of cooperative co-evolutionary genetic algorithms is highly affected by the representative selection strategy. But rational method is absent now. Oriented to the shortage, the representative selection strategy is studied based on the parallel implementation of cooperative co-evolutionary genetic algorithms in LAN. Firstly, the active cooperation ideology for representative selection and the dynamical determinate method on cooperation pool size are put forward. The methods for determining cooperation pool size, selecting cooperators and permuting cooperations are presented based on the evolutionary ability of sub-population and distributive performance of the individuals. Thirdly, the implementation steps are given. Lastly, the results of benchmark functions optimization show the validation of the method.