multi-parent genetic algorithms

multi-parent partially mapped crossover (MPPMX) and adjacency-based crossover (ABC) 可以用于解决不同combinatorial problems, 但是 MPPMX and ABC 计算时间很长或者性能很差。
 
  1. multi-parent partially mapped crossover (MPPMX) 2009 ,该算法考虑了父代基于的顺序order of the parent gene, 但是计算时间复杂度大,因为有mapping list

        参考文献:Multi-parent extension of partially mapped crossover for combinatorial optimization problems

    2.probabilistic multi-objective evolutionary algorithm (PMOEA) 2017 ,该算法考虑了父代基因出现的频率the gene frequency of parent solutions

        参考文献: A novel multi-objective evolutionary algorithm for recommendation systems

    3. (NewCross) 2020 : 同时考虑了基因的顺序order和频率frequency genes,

        参考文献:Multi-objective item evaluation for diverse as well as novel item recommendations 

    4.multi-parent order crossover (MPOX) 2019 ,考虑了基因的相对顺序

       参考文献: A novel multi-parent order crossover in genetic algorithm for combinatorial optimization problems 

 

binary coding 适用的crossover:

scanning crossover for binary coding 包括三种:uniform, occurrence based and fitness based crossover

1.Uniform scanning (U-Scan) 是uniform crossover的扩展,只生成一个子代。

这种方法使用maker标记(黑色方块),开始时候标记在第一个位置,然后每步往后移动,每次随机从父代染色体中选中一个标记基因。

2.Occurrence based scanning (OB-Scan)基于这样的前提:在父代中出现最多的基因是最好的。选择值的方法是选择大多数的那个,如果没有大多数的就选择第一个父代的那个。

但是这样其实并不好,更公平的方式是根据它们的fitness值。

3.Fitness based scanning (FB-Scan),一个基因是否被继承的概率与它所在的父代适应度值与总适应度值的比值,这种方式类似于轮盘赌。

 

4.  adjacency based crossover (ABC)是一种特殊的扫描,是为了order based 问题而设计的,如TSP。ABC与 U-Scan不同之处在于第一个基因的选择和maker更新的机制。
 
ABC子代第一个基因总是第一个父代染色体的第一个基因
 
ABC的maker选择机制:父代染色体的每一个maker都是上一次maker的下一个位置(这个位置上的基因必须是没有选择过的),此过程需要把染色体看成一个cycle。
 
 
5.diagonal crossover
 
 
Using more crossover points leads to better performance.
 
a GA using 10 parents diagonal crossover has more information before performing the selection step than a GA using the two-parents version.
 
 

real-coded 适用的crossover:

 
1.center of mass crossover (CMX)
   
 
2.multi-parent feature-wise crossover (MFX)
 
 
3.seed crossover (SX)
 
 
 
Multi-objective evolutionary federated learning (MEFL) is a machine learning approach that combines the principles of multi-objective optimization and federated learning. Multi-objective optimization is a technique that aims to optimize multiple objectives simultaneously, while federated learning is a decentralized machine learning approach that allows multiple devices to train a model collaboratively without sharing their data. MEFL is designed to overcome the limitations of traditional federated learning approaches, which often suffer from issues related to privacy, communication, and scalability. By using multi-objective optimization, MEFL can optimize the performance of the federated learning algorithm while also addressing these issues. MEFL works by dividing the optimization problem into multiple objectives, such as minimizing the loss function, reducing communication costs, and preserving privacy. A genetic algorithm is then used to optimize these objectives simultaneously, producing a set of Pareto-optimal solutions that represent the trade-offs between the different objectives. These Pareto-optimal solutions can then be used to select the best model for deployment, depending on the specific requirements of the application. MEFL has been shown to be effective in a wide range of applications, including image classification, natural language processing, and speech recognition. Overall, MEFL represents a promising approach to federated learning that can improve the privacy, communication, and scalability of the algorithm while also optimizing its performance.
评论 7
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值