A simulation-based quantitative analysis on the topological heritability of Dandelion-encoded meta-heuristics for tree optimization problems

摘要: The solutions to many optimization paradigms arising from different application domains can be modeled as a tree graph, in such way that nodes represent the variables optimized and edges evince topological relationships between variables. In these problems goal is infer an optimal graph interconnecting all under measure of fitness, for which wide portfolio exact approximative solvers have hitherto been reported related literature. this context research line interest last few years has focused on derivation solution encoding strategies suited deal with constraints imposed by configurations, particularly when encoded undergoes typical operators Evolutionary Computation. Almost contributions within area focus use standard crossover mutation Genetic Algorithms onto topology beneath individuals. However, pace at new evolutionary emerged community grown much sharply during decade. This manuscript elaborates heritability so-called Dandelion approach non-conventional operators. experimental application-agnostic-based study gravitates transmission Dandelion-encoded certain class multi-parent lie core family \((\mu +1)\) evolution particular, Harmony Search algorithm. Metrics define respect will defined evaluated over number convergence scenarios population algorithm, insightful conclusions drawn terms preserved structural properties newly produced initial population.