首页> 中文期刊> 《计算机工程与应用》 >求解0-1背包问题的量子蚁群算法

求解0-1背包问题的量子蚁群算法

         

摘要

0-1背包问题是组合优化中经典的NP难题,在蚁群算法的基础上结合量子计算提出一种求解0-1背包问题的量子蚁群算法.算法采用量子比特表示信息素,用量子旋转门来更新信息素,大量数据实例的比较测试表明,算法可有效提高蚂蚁算法的性能,减少搜索时间,具有更好的全局寻优能力.%0-1 knapsack problem is a typical NP-hard problem in combinatorial optimization. A quantum-inspired ant colony algorithm for solving the 0-1 knapsack problem is proposed which is based on the combination of ant colony optimization and quantum computing. In the algorithm,the pheromone is expressed by quantum bits,and quantum rotation gates are used to update the ant pheromone. Series of test instances validate the effectiveness of the algorithm. The proposed algorithm can reduce the searching time and has better performance in reaching the global optimum.

著录项

相似文献

  • 中文文献
  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号