Cricket Algorithm: Acoustic Attraction and Adaptive Search

Summary

The document describes the Cricket Algorithm, a population-based optimization method inspired by cricket calls and assembled from mechanisms associated with the Bat, Particle Swarm, and Firefly algorithms. Candidate solutions update velocity and position using a frequency term and movement toward the current best solution. A sound-absorption measure then selects between attraction to better candidates and a random walk near the best candidate; random noise decays over iterations to shift the search toward exploitation.

The article outlines acoustic calculations used to set frequency and attraction, gives suggested parameter settings, and reports comparative tests on optimization functions. Its stated strength is performance on medium- and high-dimensional functions, with weaker results on low-dimensional ones. These are general optimization benchmarks, not evidence of trading performance. The source notes that its implementation modifies canonical algorithms and that its description and experimental conclusions may not be fully accurate, so results should be treated as implementation-specific.

Key ideas

• The Cricket Algorithm combines frequency-based movement, attraction to the current best, and attraction to better population members.
• An absorption-based rule chooses between peer attraction and a local random walk near the best candidate.
• The algorithm reduces random perturbation over time to favor exploitation later in the search.
• The article reports stronger benchmark performance at medium and high dimensions than at low dimensions.
• Its tests concern optimization functions, so they do not establish effectiveness in trading.

Tags

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