Do Random and Chaotic Sequences Really Cause Different PSO Performance? Further Results

TL;DR

This paper investigates whether chaotic and random sequences lead to different PSO performance, finding that distribution rather than chaos or randomness primarily influences results.

Contribution

It provides a detailed analysis showing that the underlying distribution, not chaos or randomness, mainly affects PSO performance, challenging previous assumptions.

Findings

Distribution is the main factor influencing PSO performance

No systematic performance difference between chaotic and random sequences

Performance differences are due to underlying distribution, not sequence type

Abstract

Empirical results show that PSO performance may be different if using either chaotic or random sequences to drive the algorithm's search dynamics. We analyze the phenomenon by evaluating the performance based on a benchmark of test functions and comparing random and chaotic sequences according to equality or difference in underlying distribution or density. Our results show that the underlying distribution is the main influential factor in performance and thus the assumption of general and systematic performance differences between chaos and random appears not plausible.