...
首页> 外文期刊>Bulletin of the American Physical Society >APS -APS March Meeting 2017 - Event - Fitness landscape complexity and the emergence of modularity in neural networks
【24h】

APS -APS March Meeting 2017 - Event - Fitness landscape complexity and the emergence of modularity in neural networks

机译:APS -APS 2017年3月会议-活动-健身景观的复杂性和神经网络中模块化的出现

获取原文
   

获取外文期刊封面封底 >>

       

摘要

Previous research has shown that the shape of the fitness landscape can affect the evolution of modularity. We evolved neural networks to solve different tasks with different fitness landscapes, using NEAT, a popular neuroevolution algorithm that quantifies similarity between genomes in order to divide them into species. We used this speciation mechanism as a means to examine fitness landscape complexity, and to examine connections between fitness landscape complexity and the emergence of modularity.
机译:先前的研究表明,健身景观的形状可以影响模块化的演变。我们使用NEAT(一种流行的神经进化算法,用于量化基因组之间的相似性以便将它们划分为种)进化出神经网络,以解决具有不同适应性景观的不同任务。我们使用这种物种形成机制作为检验适应度景观复杂性以及检验适应度景观复杂性和模块化出现之间的联系的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 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