Turtle Shell Evolution: A Clustered Population Optimization Algorithm
Summary
The Turtle Shell Evolution Algorithm (TSEA) is a population-based optimization method that arranges candidate solutions in a shell-like structure. It groups solutions vertically by fitness and horizontally by location, with limited capacity in each cell. Better solutions occupy outer layers, while weaker candidates can remain in inner layers, preserving some diversity as the search proceeds. New candidates are produced either by perturbing selected agents or by averaging agents from selected clusters.
The procedure uses K-Means for initial grouping and nearest-neighbor assignment to place new candidates, with periodic reclustering of the vertical fitness layers. Selection favors stronger layers, while the approach also retains opportunities to draw from weaker ones. The article reports qualitative testing claims: good convergence across varied functions, alongside high result dispersion on low-dimensional functions and substantial computing demand. It does not provide enough detail here to judge performance on trading strategies, and the author notes that algorithm implementations may differ from canonical versions.
Key ideas
- TSEA groups candidate solutions by both fitness and spatial proximity in a two-dimensional shell structure.
- New candidates arise through perturbation or averaging of agents selected from shell clusters.
- K-Means initializes location clusters, and nearest-neighbor assignment places new agents into the shell.
- Selection favors higher-quality layers while retaining some weaker solutions to support exploration.
- The reported strengths and weaknesses concern optimization-function experiments, not demonstrated trading performance.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.