共计 2296 个字符,预计需要花费 6 分钟才能阅读完成。
1、A GPU accelerated Genetic Algorithm for the Construction of Hadamard Matrices
https://arxiv.org/pdf/2208.14961
Andras Balogh, Raven Ruiz
这篇论文应用遗传算法来构建 Hadamard 矩阵。生成随机矩阵的初始群体是除第一列全副是 + 1 以外,每列中都是均衡数量的 + 1 和 - 1 项。通过实现了多个适应度函数并进行筛选,找到了最无效的适应度函数。穿插过程是通过替换父矩阵种群的列来生成子代矩阵种群。渐变过程为在随机列中翻转 + 1 和 - 1 条目对。为了放慢计算速度,应用 CuPy 库在 GPU 上并行处理数千个矩阵和矩阵操作。
2、Cosmic Inflation and Genetic Algorithms
https://arxiv.org/pdf/2208.13804
Steven Abel, Andrei Constantin, Thomas R. Harvey, Andre Lukas
这是一篇对于粒子物理学和遗传算法联合的论文,我集体的了解是通过遗传算法来结构宇宙收缩的模型,这外面业余属于很多,所以贴下论文的摘要吧:
Large classes of standard single-field slow-roll inflationary models consistent with the required number of e-folds, the current bounds on the spectral index of scalar perturbations, the tensor-to-scalar ratio, and the scale of inflation can be efficiently constructed using genetic algorithms. The setup is modular and can be easily adapted to include further phenomenological constraints. A semi-comprehensive search for sextic polynomial potentials results in roughly O(300,000) viable models for inflation. The analysis of this dataset reveals a preference for models with a tensor-to-scalar ratio in the range 0.0001 < r < 0.0004. We also consider potentials that involve cosine and exponential terms. In the last part we explore more complex methods of search relying on reinforcement learning and genetic programming. While reinforcement learning proves more difficult to use in this context, the genetic programming approach has the potential to uncover a multitude of viable inflationary models with new functional forms.
3、Genetic algorithms for the resource-constrained project scheduling problem in aircraft heavy maintenance
https://arxiv.org/pdf/2208.07169
Kusol Pimapunsri, Darawan Weeranant, Andreas Riel
因为飞机衰弱治理(AHM)中的流动是互相关联并且都是大型的操作导致飞机培修停机工夫很长,许多航空公司不得对这种大量的工夫进行提前的布局。AHM 中的调度问题被认为是一个 np 难问题。应用现有算法可能是耗时的,甚至在有些状况下会产生问题。所以这篇论文提出了用于解决 AHM 中资源束缚我的项目调度问题(RCPSP)的遗传算法。这项钻研的目标是尽量缩短培修打算的竣工工夫。该算法采纳 5 条启发式调度规定,以流动列表的模式生成初始种群,采纳 RCPSP 最早开始工夫 (EST) 和工作组最早开始工夫 (WEST) 的资源分配办法对适应度值进行评估。
在抉择过程中采纳了 elitist 法和 roulette 法。而后通过穿插和渐变操作迭代改良流动列表序列。结果表明,该算法在计算工夫和资源分配方面优于现有算法
4、Quantum vs classical genetic algorithms: A numerical comparison shows faster convergence
https://arxiv.org/pdf/2207.09251
Rubén Ibarrondo, Giancarlo Gatti, Mikel Sanz
遗传算法是受达尔文进化论启发的启发式优化技术。量子计算是利用量子资源放慢信息处理速度的一种新的计算范式。因而,通过引入量子自由度来摸索遗传算法性能的潜在进步可能是将来的一个钻研方向。依照这一思路,一种模块化量子遗传算法最近被提出来,它将个体编码在独立寄存器中,该寄存器蕴含可替换的量子子程序[arXiv:2203.15039]。这篇论文对量子遗传算法和经典遗传算法进行了数值比拟,有趣味的能够看看该论文。
https://avoid.overfit.cn/post/fc347012fbab44b1b48e201d367221a5