Crow Search Algorithm for Population-Based Optimization
Summary
The article introduces the Crow Search Algorithm, a population-based method for searching an optimization space. Each candidate solution keeps a memory of its best position. During an iteration, an agent follows another agent’s remembered position by a randomized flight; if the target is aware of being followed, the pursuer instead moves to a random location. New candidates are checked against parameter bounds, evaluated, and used to update personal memories when they improve fitness. The article also outlines an implementation with population size, flight length, and awareness probability as parameters.
The method is presented as a general optimizer that could be used to tune parameters, rather than as a trading strategy. The article refers to test results and ranking plots, but the supplied text omits much of that evidence and gives no detailed comparison figures. Its stated assessment is that results have high variance and performance is poor on high-dimensional problems, despite a small parameter count and straightforward implementation. Conclusions are based on the author’s experiments, and the article notes that implementations may differ from canonical algorithms.
Key ideas
- Each crow retains its best known candidate position as personal memory.
- Agents follow another candidate with randomized movement or explore randomly when the target detects pursuit.
- Flight length and awareness probability control movement and exploration behavior.
- The article reports high result variance and weak performance on high-dimensional problems.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.