Vasileios Charilogis

A large collection of real-world classification and regression problems can be addressed using machine learning tools such as, for example, radial basis function networks (RBF networks). However, the techniques used for training RBF networks often exhibit various problems, such as getting trapped in the local minima of the error function, or even encountering numerical issues when solving systems of linear equations in order to estimate the parameters of the RBF network. This paper presents a multi-stage evolutionary technique based on genetic algorithms for the effective training of RBF networks. In the first stage, the value ranges of the RBF network parameters are estimated using the K-Means algorithm. In the second stage, the chromosomes of the genetic algorithm are initialized within the parameter ranges determined in the first stage, followed by the execution of the genetic algorithm. Each chromosome of the genetic algorithm is considered a candidate parameter vector for the machine learning model. The centers and variances of the RBF network are estimated by the genetic algorithm, while the network weights are determined by solving a system of linear equations. This method was applied to a large set of classification and data-fitting problems, yielding excellent results.

We present OptimSolution, an open-source, cross-platform framework for the systematic benchmarking and analysis of continuous optimisation methods, available for Windows, Linux and macOS and operable via both a command-line interface (CLI) and a Qt-based Graphical User Interface (GUI). The framework supports four execution modes: Single mode for single method–problem runs with convergence and distribution analysis; Batch mode for automated multi-method, multi-problem experimentation with aggregated statistical summaries; Method Sensitivity Analysis for quantifying the effect of algorithmic hyperparameters on solution quality; and Problem Sensitivity Analysis for assessing how problem-defining parameters influence instance difficulty and inter-method rankings. Complexity analysis is provided along two-axes method scalability across problem dimensionalities and problem difficulty profiles across method portfolios. Statistical validation is embedded natively through Wilcoxon signed-rank and Friedman tests with post-hoc pairwise analysis, complemented by rank tables and box-plot visualisations. All outputs are exportable as CSV files or publication-ready PNG figures. The framework is designed for extensibility: new methods and benchmark problems can be registered and activated through the GUI without modifications to the core codebase, supporting rapid experimental iteration.

This article presents a new literature-informed benchmark problem for continuous global optimization, inspired by the semantics of weather-aware irrigation scheduling. The problem is formulated as a continuous minimization task in which irrigation decisions are determined under synthetically generated but agronomically motivated weather-dependent water demand, rainfall effects, and nonlinear operational penalties. The proposed benchmark is intentionally synthetic and self-contained requiring no external datasets or field-calibrated parameters while its components are grounded in established concepts from irrigation science such as evapotranspiration-based demand estimation, yield response–water relationships, and weather-dependent scheduling principles. It is designed to be sufficiently challenging for numerical optimization while retaining a clear agronomic interpretation. The article provides the full mathematical formulation of the problem, explains its main components and parameters, and reports an experimental study focused on its optimization behavior using ten established continuous optimizers across five problem dimensions. From this perspective, the proposed problem is positioned within a clear framework for the study of real-world-inspired continuous optimization benchmarks, offering an additional case in which mathematical formulation, practical interpretation, and experimental investigation coexist in a coherent and systematic manner.

In recent years, large-scale optimization problems have become increasingly common in various fields, such as machine learning and data analysis, generating increased references to both computational cost and accurate solutions. The Grey Wolf Optimizer (GWO) is an efficient collective intelligence algorithm; however, its performance may be limited when high-permissive problems are allowed or in environments with strong multimodal landscapes. In this paper, we propose the Parallel Balanced Grey Wolf Optimizer (ParallelBGWO), a parallel extension of GWO that aims to improve the balance between exploration and exploitation of the search space. The proposed algorithm divides the total population into multiple subpopulations, which cooperate through information exchange. This cooperation reduces the probability of premature convergence and enhances the global search. The parallel implementation exploits modern multi-core computing architectures, achieving a significant reduction in execution time, while at the same time maintaining or improving the quality of the final solutions. Experimental evaluations on high-dimensional benchmark functions show that ParallelBGWO exhibits faster convergence and a reduced number of objective function evaluations compared to the classical version of GWO and other prominent methods. The results highlight ParallelBGWO as an efficient approach for demanding global optimization problems.

ARQ2 is introduced as a next-generation extension of the ARQ (Archive-guided Roulette-based with Quarantine) optimization framework, which was originally established as a cohesive strategy for achieving stable performance on difficult and noisy search landscapes. While rooted in the core principles of the original design, ARQ2 advances this foundation through a more refined integration of adaptive search, archive-guided exploration, controlled replacement, and robustness-aware population management. In doing so, it moves beyond the level of a simple algorithmic modification and emerges as a distinct methodological development within hybrid continuous optimization. Its significance lies in shaping a more mature and experimentally substantiated variant that promotes dependable behavior, consistent search quality, and balanced exploration–exploitation dynamics across diverse optimization environments. Empirical evaluation against nine established optimizers on 36 continuous optimization problems yields an overall average rank of 1.958, with 25 rank-1 placements and a mean-value improvement over ARQ on 21 out of 36 problems, confirming the stronger robustness and repeated-run reliability of the proposed design. As such, ARQ2 contributes to the ongoing development of stability-oriented optimization methodologies and reinforces the scientific relevance of this design line in the contemporary literature.

Artificial neural networks are reliable machine learning models that have been applied to a multitude of practical and scientific applications in recent decades. Among these applications, there are examples from the areas of physics, chemistry, medicine, etc. To effectively apply them to these problems, it is necessary to adapt their parameters using optimization techniques. However, in order to be effective, optimization techniques must know the range of values for the parameters of the artificial neural network, so that they can adequately train the artificial neural network. In most cases, this is not possible, as these ranges are also significantly affected by the inputs to the artificial neural network from the objective problem it is called upon to solve. This situation usually results in artificial neural networks becoming trapped in local minima of the error function or, even worse, in the phenomenon of overfitting, where although the training error achieves low values, the artificial neural network exhibits low performance in the corresponding test set. To address this limitation, this work proposes a novel two-stage training approach in which a simulated annealing (SA)-based preprocessing stage is employed to automatically identify optimal parameter value intervals before the application of any optimization method to train the neural network. Unlike similar approaches that rely on fixed or heuristically selected parameter bounds, the proposed preprocessing technique explores the parameter space probabilistically, guided by a temperature-controlled acceptance mechanism that balances global exploration and local refinement. The proposed method has been successfully applied to a wide range of classification and regression problems and comparative results are presented in detail in the present work.

Global optimization represents a fundamental challenge in computer science and engineering, as it aims to identify high-quality solutions to problems spanning from moderate to extremely high dimensionality. The Differential Evolution (DE) algorithm is a population-based algorithm like Genetic Algorithms (GAs) and uses similar operators such as crossover, mutation and selection. The proposed method introduces a set of methodological enhancements designed to increase both the robustness and the computational efficiency of the classical DE framework. Specifically, an adaptive termination criterion is incorporated, enabling early stopping based on statistical measures of convergence and population stagnation. Furthermore, a population sampling strategy based on k-means clustering is employed to enhance exploration and improve the redistribution of individuals in high-dimensional search spaces. This mechanism enables structured population renewal and effectively mitigates premature convergence. The enhanced algorithm was evaluated on standard large-scale numerical optimization benchmarks and compared with established global optimization methods. The experimental results indicate substantial improvements in convergence speed, scalability and solution stability.

Global optimization is a fundamental tool for addressing complex and nonlinear problems across scientific and technological domains. The primary objective of this work is to enhance the efficiency, stability, and convergence speed of the Magnificent Frigatebird Optimization (MFO) algorithm by introducing new strategies that strengthen both global exploration and local exploitation. To this end, we propose an improved version of MFO that incorporates three novel movement strategies (aggressive, conservative, and mixed), a BFGS-based local search procedure for more accurate solution refinement, and a dynamic termination criterion capable of detecting stagnation and reducing unnecessary function evaluations. The algorithm is extensively evaluated on a diverse set of benchmark functions, demonstrating substantially lower computational cost and higher reliability compared to classical evolutionary and swarm-based methods. The results confirm the effectiveness of the proposed modifications and highlight the potential of the enhanced MFO for application to demanding real-world optimization problems.

People with Parkinson’s disease often show changes in their movement abilities during the day, especially around the time they take medication. Being able to record these variations in an objective way can help doctors adapt treatment and follow disease changes more closely. A methodology for quantitative motor assessment is proposed in this work. It employs data from a custom SmartGlove equipped with inertial sensors. A multi-method feature selection scheme is developed, integrating statistical significance, model-based importance, and variance contribution. The most significant features were retained, and higher-level artificial features were generated using Grammatical Evolution (GE). The framework combines multi-criteria feature selection with evolutionary feature construction, providing a compact and interpretable representation of motor behavior. Additionally, the framework highlights nonlinear and composite features as potential digital biomarkers for Parkinson’s monitoring. The method was validated on recordings collected from Parkinson’s patients before and after medication intake. The recordings have been retrieved during four standardized hand motor tasks targeting tremor, bradykinesia, rigidity, and general movement anomalies. The proposed method was compared with five existing machine learning models based on artificial neural networks. GE-based features reduced classification errors to 10–19%, outperforming baseline models. Furthermore, the proposed methodology performs prediction and recall 80–88%.