multi-parent genetic algorithms
multi-parent partially mapped crossover (MPPMX) and adjacency-based crossover (ABC) 可以用於解決不同combinatorial problems, 但是
MPPMX and ABC 計算時間很長或者效能很差。
- multi-parent partially mapped crossover (MPPMX) 2020 ,該演算法考慮了父代基於的順序order of the parent gene, 但是計算時間複雜度大,因為有mapping list
- probabilistic multi-objective evolutionary algorithm (PMOEA) ,該演算法考慮了父代基因出現的頻率the gene frequency of parent solutions
- (NewCross)
- multi-parent order crossover (MPOX)
binary coding 適用的crossover:
scanning crossover for binary coding 包括三種:uniform, occurrence based and fitness based crossover
1.Uniform scanning (U-Scan) 是uniform crossover的擴充套件,只生成一個子代。
這種方法使用maker標記(黑色方塊),開始時候標記在第一個位置,然後每步往後移動,每次隨機從父代染色體中選中一個標記基因。
2.Occurrence based scanning (OB-Scan)基於這樣的前提:在父代中出現最多的基因是最好的。選擇值的方法是選擇大多數的那個,如果沒有大多數的就選擇第一個父代的那個。
但是這樣其實並不好,更公平的方式是根據它們的fitness值。
3.Fitness based scanning (FB-Scan),一個基因是否被繼承的概率與它所在的父代適應度值與總適應度值的比值,這種方式類似於輪盤賭。
4. adjacency based crossover (ABC)是一種特殊的掃描,是為了order based 問題而設計的,如TSP。ABC與
U-Scan不同之處在於第一個基因的選擇和maker更新的機制。
ABC子代第一個基因總是第一個父代染色體的第一個基因
ABC的maker選擇機制:父代染色體的每一個maker都是上一次maker的下一個位置(這個位置上的基因必須是沒有選擇過的),此過程需要把染色體看成一個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 適用的crossover:
1.center of mass crossover (CMX)
2.multi-parent feature-wise crossover (MFX)
3.seed crossover (SX)
相關文章
- 閱讀筆記(全覆蓋路徑規劃演算法):The path planning of cleaner robot for coverage region using Genetic Algorithms筆記演算法Go
- 【java】【泛型】泛型geneticJava泛型
- Deterministic Policy Gradient AlgorithmsGo
- Symbolic Discovery of Optimization AlgorithmsSymbolGo
- Study Plan For Algorithms - Part44Go
- Study Plan For Algorithms - Part45Go
- Study Plan For Algorithms - Part46Go
- Study Plan For Algorithms - Part38Go
- Study Plan For Algorithms - Part39Go
- Study Plan For Algorithms - Part40Go
- Study Plan For Algorithms - Part28Go
- Study Plan For Algorithms - Part19Go
- COMP4134 Algorithms and Data StructuresGoStruct
- Study Plan For Algorithms - Part33Go
- Study Plan For Algorithms - Part37Go
- Study Plan For Algorithms - Part29Go
- Study Plan For Algorithms - Part12Go
- Study Plan For Algorithms - Part14Go
- Study Plan For Algorithms - Part11Go
- Study Plan For Algorithms - Part13Go
- Study Plan For Algorithms - Part15Go
- Study Plan For Algorithms - Part4Go
- Study Plan For Algorithms - Part1Go
- Study Plan For Algorithms - Part6Go
- Study Plan For Algorithms - Part16Go
- Study Plan For Algorithms - Part17Go
- Study Plan For Algorithms - Part7Go
- Study Plan For Algorithms - Part9Go
- Study Plan For Algorithms - Part3Go
- Study Plan For Algorithms - Part10Go
- Analysis of Set Union Algorithms 題解Go
- Study Plan For Algorithms - Part27Go
- Study Plan For Algorithms - Part26Go
- Study Plan For Algorithms - Part25Go
- rust-algorithms:17-BWTRustGo
- XJEL1703 – Algorithms andNumerical MathematicsGo
- An Overview of High Performance Computing and Responsibly Reckless AlgorithmsViewORMGo
- rust-algorithms:16-字元逆序RustGo字元