Elite Crystal Evolution Algorithm for Population Optimization

Summary

This article presents practical components of an Elite Crystal Evolution Algorithm, a population based optimizer inspired by crystal formation. The population is divided into elite and regular agents. Elite agents conduct local search, while regular agents mix movement toward the global best, the nearest elite and the elite group's center, with random exploratory steps at different scales. A periodic wind mechanism relocates a weak non-elite agent either near the best solution or to a random point in the search space.

The document describes coordinate updates, stochastic movement weights, boundary and step-size handling, and the logic for selecting and relocating the weakest non-elite candidate. It reports comparative test results in general terms: the algorithm is described as fast and stronger on medium and high dimensional test functions, with weaker results on low dimensional functions. These are optimization benchmarks, not trading results. The author cautions that canonical algorithms may have been modified and that the conclusions reflect the experiments performed; the excerpt does not provide detailed benchmark values or trading validation.

Key ideas

• The algorithm divides a population into elite and regular candidates with different search roles.
• Regular candidates combine attraction toward elite solutions with small and large exploratory moves.
• The wind mechanism relocates the weakest non-elite candidate near the best solution or to a random location.
• Reported benchmarks favor medium and high dimensional functions, while low dimensional performance is weaker.
• The algorithm comparisons are experimental and do not establish trading performance.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.