Bison Algorithm: A Swarm Optimizer with Explorers and Elite Guidance

Summary

The article presents the Bison Algorithm, a population-based method for optimizing a continuous single-objective function. It divides candidate solutions into a large swarm group and a smaller explorer group. Swarm members move toward a target based either on a promising explorer or a weighted center of elite solutions; explorers search in a changing direction. Candidates that worsen can revert, while successful explorers can replace weaker members of the herd.

The article explains the algorithm’s configurable population, group proportions, elite size, and movement step, then discusses tests against other optimization methods. It characterizes BIA as simple and fast, but reports high variability on low- and medium-dimensional test functions and says substantial improvements would be needed to rank well in its comparison. These results concern benchmark optimization, not demonstrated trading profitability. Performance may depend on the objective landscape and parameter choices, and the method is presented as a general search technique rather than a trading strategy.

Key ideas

• BIA splits candidate solutions into a main swarm and a smaller group of explorers.
• The swarm moves toward an explorer or a weighted center of the leading candidates.
• Explorers search in changing directions and may replace weaker swarm members when they find better solutions.
• The article reports a simple, fast method but also high variability on low- and medium-dimensional benchmarks.
• Benchmark optimization results 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.