Table 2 Comparison study of IOPA with advanced optimizers.
Criterion | IOPA | BO | CMA-ES | PSO | GA | Performance |
|---|---|---|---|---|---|---|
Convergence Speed | Fast | Moderate | Moderate | Moderate | Moderate | The tuning is improved via iterative removal. |
Swarm Diversity | High (removes stagnated agents) | Moderate | Moderate | Moderate | Moderate | Exploration is maintained better than others. |
Local Minima Avoidance | Effective | Prone to local minima | Moderate | Moderate | Moderate | Local traps are evaded by iterative removal. |
Computational Overhead | Moderate (1.2–1.5× runtime) | Low | Low to Moderate | Less Effective | Less Effective | Slightly improved runtime is justified by gains. |
Accuracy Improvement | 3–5% higher | Good | Comparable | Moderate | Moderate | Better accuracy offsets overhead. |
