CAJ | 학술논문

提出一种结合变邻域搜索的离散竞争Hopfield神经网络,用于求解最大分散度问题.为了克服神经网络易陷入局部最小值的问题,将变邻域搜索的思想引入到离散竞争Hopfield神经网络中,一旦网络陷入局部最小值,变邻域搜索能帮助神经网络动态改变搜索邻域,从而跳出局部最小值去搜寻更优的解.最后,针对最大分散度问题的实验结果表明,提出的算法具有良好的性能.
제출일충결합변린역수색적리산경쟁Hopfield신경망락,용우구해최대분산도문제.위료극복신경망락역함입국부최소치적문제,장변린역수색적사상인입도리산경쟁Hopfield신경망락중,일단망락함입국부최소치,변린역수색능방조신경망락동태개변수색린역,종이도출국부최소치거수심경우적해.최후,침대최대분산도문제적실험결과표명,제출적산법구유량호적성능.
A discrete competitive Hopfield neural network (DCHNN) combined with variable neighborhood search (VNS) was proposed for the maximum diversity problem. In order to overcome the local minima problem of DCHNN, the idea of VNS was introduced into DCHNN. Once the network is trapped in local minima, the VNS can generate a new search neighborhood for DCHNN. Thus, the proposed algorithm can escape from local minima and further search better results. Finally, simulation results on the maximum diversity problem show that the proposed algorithm has good performance.