evaluate population initialization

In binary evolutionary optimization, the quality of the initial population critically affects algorithm performance under low-budget conditions (i.e., limited function evaluations), yet existing initialization methods often rely on problem-specific prior knowledge or fail to generalize across diverse real-world problems. Method: This paper proposes an adaptive, experience-based transfer initialization method that requires no problem-specific prior knowledge. It introduces a generalizable framework for representing, selecting, and transferring solution patterns—dynamically accumulating high-quality empirical knowledge from canonical benchmark problems into an experience repository—and employs a hybrid transfer strategy to adapt these patterns to unseen, complex, real-world problems and high-dimensional instances. Contribution/Results: Seamlessly integrated with standard evolutionary algorithms, the method demonstrates consistent effectiveness across six benchmark problem classes. Notably, it significantly outperforms state-of-the-art generic initialization approaches on three previously unseen real-world problems—validating its strong cross-problem generalization capability and computational efficiency under stringent evaluation budgets.

The proliferation of metaheuristic algorithms has raised concerns regarding their genuine novelty, as many purportedly innovative methods lack rigorous behavioral validation. Method: This paper systematically evaluates algorithmic distinctions from the perspective of search behavior—not performance—by conducting large-scale empirical analysis of search trajectories for 114 algorithms on standard benchmark suites using the MEALPY library. It introduces cross-matching statistical tests—a novel application in metaheuristics—for objective comparison of multivariate search distributions, enabling behavior-driven clustering and discrimination. Contribution/Results: The analysis reveals that most newly proposed algorithms exhibit search patterns highly homogeneous with classical methods, undermining claims of behavioral novelty. The study establishes a reproducible, interpretable, behavior-oriented evaluation paradigm, providing a scientific foundation for validating algorithmic design efficacy and enabling principled taxonomic classification of metaheuristics.

This study investigates the impact of population initialization methods on the performance of genetic programming for symbolic regression. Within the NSGA-II multi-objective evolutionary framework, it systematically compares three random initialization strategies against an initialization based on small-scale optimized solutions from Exhaustive Symbolic Regression (ESR) across multiple synthetic and real-world datasets. The findings reveal that although ESR-based initialization offers a modest advantage in early evolutionary stages, the choice of initialization strategy does not significantly affect the accuracy or complexity of the final Pareto front; differences between strategies vanish within a few generations. These results challenge the common assumption that sophisticated initialization substantially enhances symbolic regression performance, suggesting instead that the structure of the initial population has limited influence on long-term evolutionary outcomes.

This work addresses the limitation of conventional survival selection in evolutionary diversity optimization, which often fails due to its dependence on pairwise solution diversity. To overcome this issue, we propose a novel framework that enables the synchronous generation of multiple candidate solutions per generation, along with a tailored survival selection mechanism designed specifically for this setting. By moving beyond the traditional paradigm of single-solution, sequential updates, our approach effectively handles the dynamic nature of each solution’s contribution to population diversity. Experimental results demonstrate that, under certain conditions, the proposed multi-solution generation strategy accelerates convergence toward diverse solutions and significantly improves both the spread and quality balance of the final solution set.

Coevolutionary algorithms are widely applied in hardware design, game strategy optimization, and vulnerability repair, yet their pathological behaviors—such as gradient vanishing, relative overgeneralization, and mediocre stagnation—lead to unpredictable performance and lack rigorous theoretical guarantees. Method: This work establishes the first rigorous runtime analysis framework for population-based competitive coevolutionary algorithms, focusing on bilinear minimax optimization. It integrates probabilistic modeling, Markov chain analysis, and population dynamics to characterize convergence behavior. Contribution/Results: We precisely identify the phase transition between polynomially solvable and exponentially hard regimes: proving that a class of simple coevolutionary algorithms converges in polynomial expected time under specific conditions, while rigorously demonstrating that, with high probability, exponential time is required in other settings. This work fills a fundamental gap in the theoretical analysis of coevolutionary solvability and provides the first formal criterion for assessing algorithmic reliability.

This work proposes a population-based neural combinatorial optimization framework that addresses the limited exploratory capacity and robustness of traditional neural approaches, which typically operate on a single solution. By leveraging neural networks to jointly represent a set of candidate solutions, the framework incorporates a population-aware hierarchical classification mechanism to explicitly model inter-solution information sharing and diversity control. This design simultaneously reinforces high-quality solutions and preserves population diversity, effectively bridging the gap between neural optimization and classical population-based metaheuristics. Experimental results on the Max-Cut and Maximum Independent Set problems demonstrate that the proposed framework substantially improves both solution quality and algorithmic robustness.

This study addresses the challenge of many-objective optimization, where a large number of objectives intensifies complex interactions that critically influence algorithmic performance. To systematically investigate this phenomenon, the authors construct a diagnostic benchmark suite with controllable problem structure and objective dimensionality, enabling rigorous evaluation of prominent evolutionary algorithms—including NSGA-II, NSGA-III, MOEA/D, and lexicase selection. Experimental results demonstrate that objective interactions play a pivotal role in determining algorithm efficacy. Notably, lexicase selection is shown for the first time to match or even surpass state-of-the-art methods in many-objective settings without requiring predefined reference directions, highlighting its robustness and adaptability in high-dimensional objective spaces.

This work addresses the lack of theoretical foundations for dynamic population sizing in multi-objective evolutionary algorithms by introducing a novel bi-objective benchmark problem, CLIMB. Through rigorous runtime analysis, it compares the performance of GSEMO and NSGA-II under both fixed and dynamic population strategies. The study provides the first provable super-constant speedup of GSEMO over fixed-population NSGA-II and proposes a new variant, NSGA-II-DYN. Leveraging diversity-based evolutionary analysis, family-tree lower-bound techniques, and tools from single-objective optimization theory, the paper establishes that both NSGA-II-DYN and GSEMO converge to the Pareto front in expected $O(n \log n)$ fitness evaluations, whereas fixed-population NSGA-II requires $\Omega(n^{1.5})$, yielding an asymptotic speedup of $\Omega(\sqrt{n} / \log n)$.

Existing mainstream benchmark suites (e.g., BBOB, CEC) lack fidelity to real-world continuous and mixed-integer optimization problems—failing to reflect their structural characteristics, practical constraints, and information limitations—leading to misuse in algorithm competitions, automated algorithm selection, and industrial decision-making. Method: We propose a next-generation, scenario-driven continuous optimization benchmarking framework featuring: (1) a curated benchmark suite grounded in real-world problems; (2) an interpretable high-dimensional problem feature space coupled with an open-source performance database; (3) native support for multi-objective optimization, noisy environments, and algorithm behavioral analysis; and (4) community-driven, dynamic evolution via a collaborative platform. Contribution/Results: The framework bridges the gap between academic evaluation and industrial requirements, significantly enhancing the practicality, interpretability, and reliability of benchmarks for algorithm selection and deployment decisions. It fosters a sustainable, scientifically rigorous, and engineering-ready benchmarking ecosystem.

This work addresses the lack of effective evaluation methodologies for stopping criteria in evolutionary multi-objective optimization, a limitation that has hindered progress in the field. To this end, it proposes the first comprehensive benchmarking framework specifically designed for assessing stopping criteria. The framework introduces several innovations, including a scalar performance metric, standardized encoding of population states, and a file-based reproducible testing protocol. These components collectively enable unified, efficient, and reproducible evaluation of stopping criteria while significantly reducing storage overhead and enhancing comparative efficiency. The authors conduct systematic experiments evaluating five representative stopping criteria, demonstrating the effectiveness and practical utility of the proposed framework.