Modifying the Dingo Optimization Algorithm for Search Performance

Summary

The article presents a modified population-based optimization algorithm inspired by dingo hunting behavior. It changes three parts of the original method: group attacks move toward the best solution by changing the sign and scaling of the attack contribution; the survival procedure is reserved for agents below a much lower fitness threshold; and scavenging drops an absolute-value operation to permit movement into negative coordinates. The algorithm still uses probabilistic choices between hunting and scavenging, and between group attacks and pursuit.

The author compares the method with other optimization algorithms on test functions and reports fast execution and high average scores, alongside substantial variability and a tendency to become stuck. The article recommends using it for an initial search stage before switching to more precise methods, and stresses normalizing parameters to a common range. These results concern benchmark optimization, not evidence of profitable trading; the author also notes that implementations may differ from canonical descriptions.

Key ideas

• The modified algorithm changes group-attack direction and averages attack contributions across coordinates.
• A lower survival threshold limits the survival procedure to the weakest agents.
• Removing the absolute value from scavenging allows moves to negative coordinates.
• The reported benchmark profile combines fast search and high average results with variability and stagnation risk.
• The author suggests using the algorithm for initial exploration and following it with a more precise optimizer.

Tags

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