一种基于混沌搜索的自适应入侵遗传算法

将生物系统中“入侵”的概念引入遗传算法，提出了一种基于混沌搜索的自适应入侵遗传算法。该算法动态地引入入侵种群，并采用混沌搜索产生入侵个体。入侵种群的扩散使优良基因得以在个体间传播，优化了种群的基因构成，能够促使种群跳出局部最小，并向全局最优的方向进化，从而有效地避免了遗传算法的早熟现象。将该算法用于函数优化及解决模式分类问题的神经网络参数训练，实验结果表明，该算法具有较快的收敛速度和较强的寻优能力。

Adaptive invading genetic algorithm based on chaos search

By introducing the concept of invasion of biological systems into Genetic Algorithm (GA), a chaos-search-based Adaptive Invading Genetic Algorithm (AIGA) was proposed in this paper. The invading population, whose size was dynamically determined, was obtained through Chaos Search (CS). The expansion of the invading population was capable of propagating excellent genes among individuals and optimizing the gene structure of the population. And thus, it made the population evolve towards the global optimum. As a result, the algorithm was able to diminish the probability of being convergent to local minima prematurely. The proposed algorithm was applied to function optimization and pattern classification. And the experimental results show that this algorithm has the merits of fast convergence and global optimization.