Objective
To construct an efficient optimizer capable of maintaining search diversity in early iterations while preserving stable convergence pressure in later iterations.
Metaheuristic Optimization Study
KING: An Efficient Optimization Approach
KING formulates a stage-wise population evolution strategy that couples adaptive confrontation, Levy-assisted local refinement, and consensus-guided convergence for nonlinear global optimization.
Structured Abstract
A concise research-oriented summary of the formulation and empirical behavior of KING.
Method
KING integrates staged interactions among candidate solutions through elite guidance, randomized confrontation, and adaptive control of search power.
Results
Reported experiments indicate that KING demonstrates competitive performance on benchmark optimization suites and practical optimization tasks.
Conclusion
The algorithm offers a balanced and robust search mechanism with reduced premature convergence risk, suitable for complex continuous optimization scenarios.
Methodological Framework
Four coordinated stages are employed to model the transition from global exploration to refined exploitation.
Ascent of Might
The initialization stage evaluates candidate quality and calibrates adaptive search pressure, establishing a principled trajectory for exploration intensity.
Joint Confrontation
Candidate updates are generated by weighted interactions among elite references, current states, and random perturbations, expanding feasible-domain coverage.
Three-Legged Tripod
Local refinements combine better-solution guidance with Levy-driven motion, enabling effective escape from deceptive local optima.
Whole Country United
Consensus updates around high-quality solutions reinforce convergence reliability, while boundary absorption preserves numerical feasibility.
Algorithmic Highlights
Stage Coupling
Sequentially coupled operators support a smooth exploration-to-exploitation transition.
Search Balance
Adaptive power control and confrontation dynamics regulate diversification and intensification.
Convergence Stability
Consensus refinement improves late-stage convergence behavior across complex landscapes.
Resources
Visual Materials
Zhao D, Wang Z, Li Y, Heidari A A, Wu Z, Chen Y, Chen H.
KING: An Efficient Optimization Approach.
Neurocomputing, 2025: 131645.