
This archive contains a technical preprint introducing Hierarchical Evolution with Lineage Incubation (HELI), a mechanism for stabilising structural mutation in neuroevolution. Structural mutations such as adding or removing neurons or connections often cause immediate fitness degradation and are therefore eliminated before their potential benefits can emerge. HELI addresses this issue by temporarily isolating structurally mutated individuals into short-lived lineage populations. Each lineage is evolved independently for a limited number of generations, allowing local recovery before the best individual is reintegrated into the main evolutionary population. The paper defines the HELI mechanism, describes its algorithmic integration into a standard evolutionary loop, and evaluates its behaviour on small, deterministic diagnostic benchmarks under evaluation-budget normalisation. The study focuses on controlled, reproducible analysis of structural mutation dynamics rather than large-scale or state-of-the-art performance. The HELI mechanism is implemented in the open-source EvoLib framework.
lineage incubation, structural mutation, neuroevolution, evolutionary algorithms, neural network topology
lineage incubation, structural mutation, neuroevolution, evolutionary algorithms, neural network topology
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
