论文部分内容阅读
量子进化算法采用多个简单概率模型并行搜索的框架结构,从而可尝试引入有效的多模型学习机制以提高算法的探索能力.文中将全面学习的思想引入多量子概率模型的学习,提出基于全面学习的量子分布估计算法.在该算法中,模型的每个分量都可以向不同的目标学习,使得量子概率模型有可能较为全面地从已知较优解中提取知识,以尽可能全面地描述解空间中好的区域,有效提高算法求解复杂优化问题的能力.在典型0-1背包问题上的比较实验充分验证该算法的有效性和先进性.
The quantum evolutionary algorithm uses the frame structure of multiple parallel searches of probabilistic models, so that an effective multi-model learning mechanism can be introduced to improve the exploration ability of the algorithm.Introducing the idea of comprehensive learning into the study of multiple quantum probabilistic models, In this algorithm, each component of the model can be learned from different targets, making it possible for the quantum probability model to extract knowledge from the known optimal solutions more fully to describe the solution as completely as possible The good region in space can effectively improve the ability of the algorithm to solve complex optimization problems.Comparison experiments on a typical 0-1 knapsack problem fully verify the validity and advancement of this algorithm.