ECTA 2025 Abstracts

Area 1 - Theory and Methods
Full Papers Short Papers
Area 2 - Applications
Full Papers Short Papers

Area 1 - Theory and Methods

Full Papers
Paper Nr: 13
Title:

Genetic Programming with Ranging-Binding Mechanism for Symbolic Regression
Authors:

Wen-Zhong Fang, Chi-Hsien Chang, Jung-Chun Liu and Tian-Li Yu
Abstract: This paper presents ranging-binding genetic programming (RBGP), a novel algorithm for symbolic regression. RBGP uses both syntax and semantics to detect hidden structural patterns and preserves advantageous traits from parent programs during evolution. The algorithm consists of two main components: binding and ranging. The binding mechanism detects frequently occurring two-layer substructures and prevents their disruption during recombination, thus retaining essential building blocks. The ranging mechanism modifies the output range of subtrees through tree replacement to align with the target value range. We evaluate RBGP against several state-of-the-art and widely used genetic programming methods, including GP-GOMEA, ellynGP, ellynGP with epsilonlexicase selection, ellynGP with age-fitness Pareto optimization, and gplearn. Experiment results show that RBGP achieves the lowest average mean absolute error on 30 out of 37 benchmarks from the Penn machine learning benchmark, which demonstrates its generalization across diverse problem domains. An ablation study confirms the individual contributions of the binding and ranging mechanisms.
Paper Nr: 30
Title:

Scaling Up Pareto Local Optimal Solutions Networks: Modelling Multi-Objective Landscapes
Authors:

Gabriela Ochoa, Hernan Aguirre, Arnaud Liefooghe and Sébastien Verel
Abstract: Pareto local optimal solution networks (PLOS-nets) model fitness landscapes as graphs where nodes are Pareto local optima and edges account for standard neighbourhood relationships. PLOS-nets capture the global structure of multi-objective combinatorial landscapes and provide a rich set of features. Published results so far focus on small problems. This article proposes new modelling and sampling methodologies to scale up PLOS-nets to larger problem sizes. Our sampling approach is based on the effective multi-objective random one-bit climber (moRBC). In our coarse-grained model, nodes are sets of Pareto local optimal solutions grouped according to the logarithm of their Pareto rank (following non-dominated sorting). For modelling edges, we consider the original neighbourhood edges, but also propose a new definition representing softrestarts from Pareto local optima. We analyse and visualise our models on a set of combinatorial landscapes with tuneable ruggedness and number of objectives (MNK-landscapes). The models provide landscape features that correlate with the performance of multi-objective optimisation algorithms, and can be used in algorithm recommendation settings. The model with neighbour edges slightly outperforms the soft-restart model (in terms of predictive power), but its substantial computational overhead makes the soft-restart model a promising approximation.
Paper Nr: 62
Title:

An Adaptive Redundancy-Aware Binary Grey Wolf Optimizer for Feature Selection
Authors:

Asma Amdouni
Abstract: Feature selection is indispensable for reducing dimensionality and improving performance and interpretability in high-dimensional domains such as gene expression analysis. Traditional binary adaptations of the Grey Wolf Optimizer (BGWO) often suffer from premature convergence and tend to select feature subsets containing high redundancy when relying on simple single-objective fitness functions. To overcome these limitations, we introduce the Adaptive Redundancy-aware Binary Grey Wolf Optimizer (AR-BGWO), a novel single-objective metaheuristic specifically tailored for binary feature selection. AR-BGWO incorporates two principal innovations: a non-linear, stagnation-responsive adaptation of the exploration-exploitation parameter a, which enables more effective navigation of the search space, and an implicit redundancy penalty within the fitness function that discourages highly correlated feature selections, thereby promoting both accuracy and conciseness without resorting to multi-objective formulations. Experimental analyses indicate that ARBGWO consistently identifies smaller, less redundant feature sets while achieving higher or comparable classification accuracy relative to standard BGWO and other state-of-the-art methods.
Paper Nr: 102
Title:

Extending Cartesian Genetic Programming via Iterative Subgraph Assessment
Authors:

Henning Cui, Camilo De La Torre, Sylvain Cussat-Blanc, Hervé Luga, Dennis G. Wilson and Jörg Hähner
Abstract: Cartesian Genetic Programming (CGP) is a graph-based evolutionary representation in which candidate solutions are encoded as directed acyclic grid of computational nodes. In standard CGP, only the output of the full graph is considered for fitness evaluation, although all intermediary (active) node outputs are computed during execution. We introduce Iterative Subgraph Assessment CGP (ISA-CGP), a straightforward extension that treats every active active node output as a potential solution: during each individual’s evaluation, all subgraph outputs are assessed alongside the full graph, and the best-performing expression is selected. Favourably, in Symbolic Regression (SR) fitness measurements are usually inexpensive. To validate ISA-CGP, we conduct experiments on eight benchmark problems drawn from the Feynman symbolic regression suite, comparing convergence speed, final model error, and computational effort against standard CGP. Experimental results on the eight Feynman problems demonstrate that ISA-CGP converges more rapidly and attains superior fitness values in most cases. Furthermore, ISA-CGP creates smaller solution programs, indicating less bloated phenotypes which saves computational efforts. These findings suggest that ISA-CGP offers a simple yet effective enhancement to CGP, achieving faster search and better solutions with minimal overhead.
Paper Nr: 124
Title:

Evolution of Wavelets and Hardware for Satellite Image Compression
Authors:

Allyn Loyd and Jason A. Yoder
Abstract: When designing a system to transmit satellite images, it is important to balance the need for a high compression rate with the ability to accurately reconstruct them. The discrete wavelet transform (DWT) can be used to compress images and can be fine-tuned for specific classes of images, but it requires substantial hardware resources to perform floating-point multiplications. To increase the efficiency of this process, some efforts have been made to evolve solutions which may not follow tradition human design principles and yet arrive at more efficient solutions. Specifically, separate research efforts to evolve wavelets and hardware multipliers have been performed, but not together. This work explores the use of several evolutionary and coevolutionary algorithms to address this gap. Chosen algorithms include an evolutionary approach that evolves the wavelets and hardware in two separate stages, an evolutionary algorithm that evolves wavelets and hardware in pairs, and a cooperative coevolutionary algorithm that evolves populations of wavelets and hardware. Solutions are evaluated based on the error in image reconstruction, the accuracy of the multipliers, and the number of components used by the evolved multipliers. The results indicate that problem decomposition is most beneficial when evolution allows individual components to be optimized separately. Additionally, our evolved wavelets show improved accuracy over existing wavelets (in JPEG2000) at high compression ratios, and the evolved hardware uses four times fewer components as a human-designed approach. This work demonstrates strong potential for effective image compression and, more broadly, the use of coevolutionary algorithms to fine-tune circuit design alongside secondary systems defining the needs of those circuits (wavelet coefficients in our case).
Paper Nr: 156
Title:

A Surrogate-Assisted Co-Evolutionary Framework for Bilevel Optimization
Authors:

Sanup Araballi, Venkata Gandikota, Pranay Sharma, Prashant Khanduri and Chilukuri K Mohan
Abstract: Bilevel optimization, a class of hierarchical optimization problems with broad applications in machine learning and engineering, presents a significant research challenge due to its inherent NP-hard nature in the non-convex optimization problems. In this paper, we address the limitations of existing solvers. Classical, gradient-based methods are inapplicable to the non-convex and non-differentiable landscapes, common in practice. On the other hand, derivative-free methods like nested evolutionary algorithms are rendered intractable by a prohibitively high query complexity. To this end, we propose a novel framework, the Surrogate Assisted Co-evolutionary - Evolutionary Strategy (SACE-ES), which synergizes the global search capabilities of evolutionary computation with the data-driven efficiency of surrogate modeling. The core innovation of our framework is a multi-surrogate, constraint-aware architecture that decouples the complex bilevel problem by using separate Gaussian Process models to approximate the lower-level optimal solution vector and its corresponding constraint violations. We conducted a comprehensive empirical study on a suite of challenging benchmarks, including the standard SMD problems. We demonstrate empirically that on complex constrained problems, SACE-ES discovers solutions that are upto two orders of magnitude superior upper-level fitness values to those found by exhaustive search baselines. Furthermore, on a range of unconstrained, non-convex problems, our approach achieves statistically comparable solution quality while reducing the required computational cost by approximately upto 96%. Our results establish SACE-ES as a robust and highly efficient framework that makes previously intractable bilevel problems solvable, particularly in the presence of complex constraints.
Paper Nr: 163
Title:

SA-GA: Surrogate-Assisted Genetic Algorithm for Optimizing Activation Function Approximations in Privacy-Preserving EEG Classification Networks
Authors:

Essam Debie and Harshita
Abstract: Electroencephalography (EEG) data is highly sensitive, requiring strong privacy protection in brain–computer interfaces and neurological diagnosis. Homomorphic encryption enables secure EEG classification with deep neural networks, but standard activation functions (e.g., ReLU) are incompatible with encrypted arithmetic, causing severe computational overhead. Polynomial approximations mitigate this issue yet fail to capture EEG’s distinctive spectral patterns, temporal dynamics, and inter-subject variability. We propose a surrogate-assisted genetic algorithm (SA-GA) to optimize activation function approximations for encrypted deep learning models. SA-GA uses fixed-length chromosome encoding and a Gaussian Process surrogate to efficiently search the approximation space while avoiding costly encrypted evaluations. Applied to motor imagery classification (BCI Competition IV Dataset 2a), our method achieves 89.7% accuracy—7.2% higher than conventional polynomial approaches—while remaining feasible for real-time use. This enables practical, privacy-preserving deployment of EEG deep learning systems in clinical and BCI applications.
Short Papers
Paper Nr: 11
Title:

Evolutionary Multi-Objective Optimization of Unmanned Aerial Base Station Placement for Power Grid Recovery
Authors:

Mohammad Reza Mohebbi, Anna Volkova and Hermann de Meer
Abstract: Rapidly restoring interconnected Information and Communication Technology (ICT) and power systems is crucial in major disasters. Advanced procedures involve communication between distributed power grid assets to form and synchronize self-contained grid islands at various voltage levels. However, disruptions in Information and Communication Technology (ICT) infrastructure can result in unrecoverable nodes, causing partial restoration and segmenting the power grid into isolated islands. This paper proposes an algorithm to optimally position Unmanned Aerial Vehicle Base Stations (UAV-BSs) to improve communication between these isolated grid islands. To use the minimum number of Unmanned Aerial Vehicles (UAVs) to reconstruct the network, it is necessary to employ a suitable algorithm to determine their positions. In this article, an algorithm is proposed to determine the position of UAVs using the minimum number of UAVs, while also considering other important goals in network reconstruction. Due to the multi-objective nature of the proposed algorithm, the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) basic algorithm has been used. According to the conditions of the problem and to improve the proposed algorithm in terms of time and to avoid finding the local optimum, the initial population of the NSGA-II has been created using clustering algorithms. The evaluation is done on synthetic scenarios and real network models. The results show that a placement strategy can find a trade-off between the objectives of network recovery and communication quality, and the proposed initialization clustering improves the algorithm performance.
Paper Nr: 12
Title:

A Constraint-Handling Method for Model-Building Genetic Algorithm: Three-Population Scheme
Authors:

Yu-Hao Kao, Chi-Hsien Chang and Tian-Li Yu
Abstract: To solve constrained optimization problems (COPs) with genetic algorithms, different methods have been proposed to handle constraints, but none of them are specifically designed for model-building genetic algorithms (MBGAs). This paper presents a three-population scheme, abbreviated as B-3Pop, which features three populations: a feasible one, an infeasible one, and a third one to explore the boundary between the feasible and infeasible spaces with MBGAs. The core idea is to learn how to combine feasible and infeasible solutions to evolve optimal solutions near the boundary. Empirically, B-3Pop outperforms five widely used constraint-handling methods—elimination, dominance concept, penalization, adaptive segregational constraint-handling methods, and the feasibleinfeasible two-population scheme—in terms of the number of function evaluations on all six tested COPs: dimensional knapsack, uncapacitated warehouse location, Steiner tree, capacitated minimum spanning tree, capacitated p-median, and weighted maximum-2-satisfiability problems.
Paper Nr: 23
Title:

Comparative Analysis of Performance Predictors in Multi-Objective Neural Architecture Search for Image Super-Resolution: XGBoost Regressor and SynFlow
Authors:

Sergio Sarmiento-Rosales, Jesus Leopoldo Llano-García, Jesper Michiel Janssen, Jesús Guillermo Falcon-Cardona, Raúl Monroy and Víctor Adrián Sosa-Hernández
Abstract: Single Image Super-Resolution reconstructs high-fidelity images from low-resolution inputs under tight computational budgets, essential in fields like medical diagnostics, remote sensing, surveillance, and aerospace imaging. To ease manual network design, Neural Architecture Search is a key AutoML tool, though its effectiveness hinges on often costly and imperfect performance evaluations. This paper compares two efficient predictors: (i) SynFlow, a zero-cost method, and (ii) an XGBoost regressor, a model-based approach. Both are integrated into NSGA-III, a multi-objective evolutionary algorithm, to explore SuperResolution architectures optimizing three objectives: maximizing PSNR while minimizing parameter count and floating-point operations. We analyze each method’s impact on final architecture quality. SynFlow offers faster and more reliable estimations compared to XGBoost, which is slower and less accurate. However, XGBoost enables broader approximations of the Pareto front, making it suitable for more comprehensive trade-off exploration. Therefore, we recommend using SynFlow in scenarios where quick and resource-constrained searches are required, while XGBoost is more appropriate when the benefits of a wider trade-off analysis outweigh the additional computational cost.
Paper Nr: 32
Title:

Comparative Analysis of Model Selection Criteria for Symbolic Regression Using Genetic Programming
Authors:

Fitria Wulandari Ramlan, Gabriel Kronberger, Colm O'Riordan and James McDermott
Abstract: Symbolic regression (SR) using genetic programming (GP) can generate a diverse set of candidate models that balance accuracy and complexity, particularly when configured with multi-objective optimisation, which produces a Pareto front of non-dominated solutions. However, selecting a single model from this population remains challenging, especially when relying only on training data. This study evaluates the effectiveness of model selection criteria in SR, which include Mean Squared Error (MSE), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Description Length (DL), and PySR Score Metric (PSM). These criteria are evaluated using their training scores on 20 realworld regression datasets from the PMLB collection using PySR. We calculate the Spearman rank correlation coefficient (ρ) between each metric and test MSE to evaluate how well the metric ranks generalisable models. The results show that no single metric performs reliably across all datasets. Metrics that focus mainly on accuracy often lead to overfitting, while simplicity-based metrics can underfit. PSM aims to balance accuracy and complexity, but its performance is inconsistent, sometimes helpful, but often unstable across datasets. This study provides practical insights into the behaviour of model selection metrics in SR and offers guidance for selecting models that generalise well without overfitting.
Paper Nr: 41
Title:

Automatic Synthesis of Selection Operators in Genetic Algorithms Using FunSearch
Authors:

Aleksandra Korableva and Vladimir Stanovov
Abstract: This paper presents an approach for the automatic synthesis of a rankbased selection operator in a binary genetic algorithm using the FunSearch method, which combines the capabilities of large language models and evolutionary search. Instead of manually designing the rank selection scheme, the algorithm autonomously discovers an implementation of the ranking function. The developed algorithm is tested on pseudo-Boolean optimization problems using 17 functions from the IOHprofiler benchmark suite. The results show that the automatically discovered operators achieve higher average fitness and stability compared to the classical scheme.
Paper Nr: 64
Title:

Enhanced Adaptive Differential Evolution for Optimized Ensemble Learning in Diabetic Retinopathy Diagnosis
Authors:

Amani Trad, Olfa Fakhfakh and Ghaith Manita
Abstract: Diabetic retinopathy (DR) remains a leading cause of vision impairment worldwide, necessitating early and precise detection to prevent irreversible damage. To this end, the present study introduces an adaptive ensemble learning framework that integrates transfer learning with an Enhanced Adaptive Differential Evolution (EADE) algorithm for robust DR classification. The ensemble comprises three convolutional neural networks—DenseNet121, Xception, and InceptionResNetV2—pre-trained on ImageNet and fine-tuned on retinal fundus images, whose outputs are aggregated via an Enhanced Weighted Voting (EWV) scheme. Ensemble weights are optimized by EADE, which extends the JADE algorithm through adaptive control of the mutation factor and crossover rate and the incorporation of an external archive to preserve population diversity. Evaluation using synthetic performance metrics indicates that the EADE-driven ensemble surpasses static weighting approaches and canonical Differential Evolution in terms of classification accuracy, macro-averaged F1-score, and ROC AUC. These findings suggest that the proposed EADE–EWV framework offers a scalable, effective decision-support tool for automated DR screening.
Paper Nr: 72
Title:

SCOPE for Hexapod Gait Generation
Authors:

Jim O'Connor, Jay B. Nash, Derin Gezgin and Gary P. Parker
Abstract: Evolutionary methods have previously been shown to be an effective learning method for walking gaits on hexapod robots. However, the ability of these algorithms to evolve an effective policy rapidly degrades as the input space becomes more complex. This degradation is due to the exponential growth of the solution space, resulting from an increasing parameter count to handle a more complex input. In order to address this challenge, we introduce Sparse Cosine Optimized Policy Evolution (SCOPE). SCOPE utilizes the Discrete Cosine Transform (DCT) to learn directly from the feature coefficients of an input matrix. By truncating the coefficient matrix returned by the DCT, we can reduce the dimensionality of an input while retaining the highest energy features of the original input. We demonstrate the effectiveness of this method by using SCOPE to learn the gait of a hexapod robot. The hexapod controller is given a matrix input containing time-series information of previous poses, which are then transformed to gait parameters by an evolved policy. In this task, the addition of SCOPE to a reference algorithm achieves a 20% increase in efficacy. SCOPE achieves this result by reducing the total input size of the time-series pose data from 2700 to 54, a 98% decrease. Additionally, SCOPE is capable of compressing an input to any output shape, provided that each output dimension is no greater than the corresponding input dimension. This paper demonstrates that SCOPE is capable of significantly compressing the size of an input to an evolved controller, resulting in a statistically significant gain in efficacy.
Paper Nr: 78
Title:

Novel Discretization Scheme for Multidimensional Split-on-Demand on Real-Valued Optimization with High Multi-Modality
Authors:

Che Wei Liang and Tian-Li Yu
Abstract: Real-valued optimization has attracted considerable interest because of its wide-ranging applications in areas such as control system design, circuit design, and hyperparameter tuning for machine learning. In this paper, we propose the skew multidimensional split-on-demand (smSoD), an extension of multidimensional split-on-demand (mSoD). mSoD serves as a discretization interface that enables discrete model-building genetic algorithms to tackle continuous-domain problems; smSoD further improves split-point selection by leveraging the sample distribution. We then embed smSoD into the integer version of the extended compact genetic algorithm (ECGA) and evaluate it on two benchmark suites: decomposable linkage problems and an extended version of the CEC2014 benchmark with additional subproblems. We compare smSoD+ECGA against both mSoD+ECGA and L-SHADE. According to statistical tests, under high-dimensional scenarios, our method outperforms the other two methods on more than 50% of the test problems.
Paper Nr: 113
Title:

Sustainable Performance Improvement of Surrogate-Assisted Evolutionary Algorithms Using Tabu Search
Authors:

Kei Nishihara, Masaya Nakata and Shinya Watanabe
Abstract: Surrogate-assisted evolutionary algorithms (SAEAs), also known as data-driven evolutionary optimization algorithms, are a representative approach for expensive optimization problems, as solutions for expensive function evaluations (FEs) are prescreened or iteratively improved with the surrogate model of objective functions. These days, SAEAs are often designed to enhance convergence speed by utilizing previously evaluated solutions with good function values. Although this convergence-first design enables SAEAs to find relatively better solutions quickly at the beginning of the search, SAEAs tend to select quite similar solutions to each other for expensive FEs in the middle to the end of the search, resulting in the waste of expensive FEs. Accordingly, this work first incorporates a principle of tabu search into existing SAEAs as well as observes the behavior of the existing SAEAs and their extension. Tabu search extended for continuous optimization prevents to evaluation of similar solutions to previously evaluated ones. This promotes a variety of solution choices in SAEAs and realizes a sustainable performance improvement toward the end of the search. In the experiments held in this work, some SAEAs that have different search strategies and machine learning models for surrogate models are selected, and their extended versions incorporating tabu search are tested in a suite of single-objective continuous optimization benchmarks under an expensive scenario.
Paper Nr: 116
Title:

A Comparative Study of Binary-Coded Red Deer and Genetic Algorithms
Authors:

Ethan Padden and Colm O'Riordan
Abstract: This study presents a comparative analysis of a binary-coded Red Deer Algorithm (RDA) and a binary-coded Genetic Algorithm (GA) across a diverse suite of benchmark problems. The RDA, a nature-inspired metaheuristic based on the mating behaviour of red deer, has shown promise in continuous optimisation but remains underexplored in binary search spaces. We implement and evaluate a binary-coded RDA (bRDA) on both synthetic and real-world combinatorial benchmarks. Performance is assessed not only in terms of final solution quality and convergence dynamics, but also through population diversity analysis over time, offering insights into how each algorithm balances exploration and exploitation throughout the search. Unlike prior studies that report performance using generation count or runtime, we present all results as a function of the number of fitness evaluations, allowing for a fair, implementation-independent comparison. Our findings show that while the GA excels in assembling coherent building blocks, the bRDA maintains greater population diversity early in the search and adapts more effectively to deceptive and heterogeneous landscapes. These results underscore the importance of diversity-aware design and suggest that hybrid algorithms may benefit from combining the GA's efficient building-block assembly with the RDA's local refinement.
Paper Nr: 117
Title:

An Overview of Genetic Algorithms in Educational Research from 2020 to 2024
Authors:

Annamaria De Santis and Tommaso Minerva
Abstract: Genetic Algorithms (GAs) applications can handle the complexity of educational phenomena and learning/teaching processes in searching for optimal solutions, managing numerous variables, and dealing with uncertainty and nonlinear problems frequently in the social sciences field. This paper examines the use of GAs in educational research from 2020 to 2024, aiming to identify current applications and new opportunities. The overview comprises 67 articles and proceeding papers retrieved from Scopus and WoS. During the investigation process, we combined traditional and AI-based tools to identify relevant areas and categorize the selected documents. The explored elements include the topics of the studies, disciplinary sectors, level of education, and the contemporary use of other techniques in conjunction with GAs.
Paper Nr: 141
Title:

Integral Maude Operation Semantics for Algebraic Petri Nets: An Effective Model of Adaptive Systems
Authors:

Lorenzo Capra
Abstract: Traditional and high-level Petri nets are limited in representing adaptive or evolving distributed systems. We introduce a comprehensive definition of Reisig’s Algebraic Petri nets (APN) with active tokens, employing Maude, a declarative language with rewriting logic semantics. Reisig’s APN offers unmatched expressivity and analytical power, and with active tokens, they represent distributed components with internal logic naturally. Active tokens facilitate straightforward and efficient meta-modeling. Although rewriting logic was proposed as a unified logical framework for PNs two decades ago, ours is the first complete implementation of APNs (with active tokens) based on Maude. We tackle modeling challenges from Maude’s rewriting strategy relying on pattern-matching and coherence by proposing two alternative definitions. Our approach is illustrated with code snippets and examples, featuring an advanced model of an adaptive MLFQ algorithm.
Paper Nr: 166
Title:

Solving the Electric Vehicle Routing Problem with Nonlinear Charging Functions Using Genetic Programming
Authors:

Magda Smolić-Ročak, Marko Đurasević and Josip Hrvatić
Abstract: The Electric Vehicle Routing Problem (EVRP) has gained increasing attention with the growing adoption of electric vehicles, driven by the global shift towards reducing the negative environmental impact. This paper addresses the EVRP with Time Windows (EVRPTW), incorporating nonlinear charging functions that are often overlooked in existing models. Traditional EVRP approaches typically assume linear charging functions, which can lead to imprecise solutions due to unrealistic charging time estimations. However, real-world charging behaviour is nonlinear, and since recharging significantly affects total travel time, it is essential to account for more realistic charging dynamics. The objective of this paper is to solve the EVRPTW while minimizing both the total number of vehicles and the overall delay, using a more accurate representation of the charging process. To further assess adaptability, several recharging policies are examined, including minimal recharging, fixed-level recharging, and an adaptive policy that adjusts based on vehicle load. The proposed approach uses genetic programming as a hyper-heuristic to evolve routing policies that construct solutions while accounting for nonlinear charging behaviour.
Paper Nr: 51
Title:

Hierarchical Self-Evaluated Topology PSO with Grouping and Mediator
Authors:

Kang Yin and Yuji Sato
Abstract: Particle Swarm Optimization (PSO) performance relies heavily on interaction topology. Our prior work, Self-Evaluated Topology Particle Swarm Optimization (SET-PSO), achieved high accuracy and stability through self-evaluated dynamic topology, but suffered from significant computational inefficiency due to global topological changes and multi-run evaluations. To solve this problem, this paper proposes the Grouping and Mediator Self-Evaluating Topology Particle Swarm Optimization (GMSET-PSO). GMSET-PSO incorporates a novel Grouping and Mediator model, organizing particles into localized Von Neumann groups with virtual mediator particles facilitating hierarchical, inter-group communication. This design effectively blocks global information propagation, localizing search and reducing computational burden. Furthermore, dynamic topology adaptation becomes group-specific, and topology evaluation is streamlined to a single-run assessment with immediate rollback if no improvement. On the CEC2020 and CEC2022 benchmark function set, GMSET-PSO was compared with SET-PSO and other PSO variants equipped with adaptive mechanisms. The results show that GMSET-PSO largely retains the accuracy of the original algorithm while improving stability by 46.08% and efficiency by 25.90%. Moreover, it significantly outperforms the other algorithms, particularly in solving more complex problems. Showcasing a significant advancement in scalable and efficient self-adaptive optimization.
Paper Nr: 75
Title:

Niching Agents in The Core
Authors:

Gary B. Parker, Jim O'Connor and John Asaro
Abstract: The Core is a unique competitive co-evolution algorithm that allows agents to evolve autonomous control without utilizing a traditional fitness function. The agents evolve via local interactions through tournament selection, crossover, and mutation, producing offspring by evolving better controllers. Previous works have shown The Core’s ability to evolve agents capable of combat and navigation in the Xpilot video game. This research expands upon that premise by niching agents to specific subsets of the original environment The Core was tested in. Our results demonstrate the niched agents capacity for success over agents niched to the entire system and agents niched to different subenvironments.
Paper Nr: 76
Title:

Decentralized Evolution of Hexapod Gaits with Independent Leg Controllers
Authors:

Gary B. Parker, John Asaro and Jim O'Connor
Abstract: This paper presents a novel approach to hexapod locomotion by evolving each leg’s gait independently through a decentralized evolutionary algorithm. Using the Webots simulator and the Mantis hexapod robot, we optimize individual leg controllers without centralized coordination, allowing emergent behaviors to drive the development of efficient, coordinated locomotion. Our decentralized method is benchmarked against cooperative coevolution, demonstrating improved efficacy in generating stable and adaptive gaits while showing interesting emergent coordination. By enabling independent evolution of leg controllers, this method reduces the complexity of gait optimization and highlights the potential of decentralized strategies for scalable and adaptive robotic systems.
Paper Nr: 79
Title:

Investigation of Behavioral Cloning Guided Genetic Programming Using a Multilayer Perceptron for Symbolic Regression
Authors:

Liang-Wei Lee and Tian-Li Yu
Abstract: This paper proposes a genetic programming (GP) algorithm for symbolic regression (SR), called the behavioral cloning guided genetic programming (BCGP) algorithm. The goal is to improve the effectiveness of crossover by preserving the relationship between parent operators and subtrees through imitating the behavior of subtree crossover during evolution. Specifically, BCGP investigates the application of a multilayer perceptron to capture features based on the parent operators and the subtrees. Across the benchmark problems from the SR benchmark and the Feyman SR database, BCGP gains the lowest count of maximum average MAEs and lowest average ranks, and statistically significantly outperforms ellynGP, and GP-GOMEA on 31 and 9 benchmark problems, respectively, out of a total of 43.
Paper Nr: 110
Title:

Robustness of Evolved Strategies: An Application to Cycling
Authors:

Donal Kelly and Colm O'Riordan
Abstract: This paper explores the evolution of strategies for teams participating in track cycling. The performance of a team depends on their strategy. Evolutionary algorithms can help find good strategies to help teams improve their times. However, questions arise about the robustness of the strategies. Strategies that lead to good times may be more fragile, leading to huge changes in performance if they are not followed perfectly. Strategies exhibiting good performance and robustness may be more desirable. We explore the robustness of evolved strategies and present results for different team configurations.
Paper Nr: 112
Title:

Behaviour Space Analysis of LLM-Driven Meta-Heuristic Discovery
Authors:

Niki van Stein, Haoran Yin, Anna V. Kononova, Thomas Bäck and Gabriela Ochoa
Abstract: We investigate the behaviour space of meta-heuristic optimization algorithms automatically generated by Large Language Model driven algorithm discovery methods. Using the Large Language Evolutionary Algorithm (LLaMEA) framework with a GPT o4-mini LLM, we iteratively evolve black-box optimization heuristics, evaluated on 10 functions from the BBOB benchmark suite. Six LLaMEA variants, featuring different mutation prompt strategies, are compared and analysed. We log dynamic behavioural metrics including exploration, exploitation, convergence and stagnation measures, for each run, and analyse these via visual projections and network-based representations. Our analysis combines behaviour-based projections, Code Evolution Graphs built from static code features, performance convergence curves, and behaviour-based Search Trajectory Networks. The results reveal clear differences in search dynamics and algorithm structures across LLaMEA configurations. Notably, the variant that employs both a code simplification prompt and a random perturbation prompt in a 1+1 elitist evolution strategy, achieved the best performance, with the highest Area Over the Convergence Curve. Behaviour-space visualizations show that higher-performing algorithms exhibit more intensive exploitation behaviour and faster convergence with less stagnation. Our findings demonstrate how behaviour-space analysis can explain why certain LLM-designed heuristics outperform others and how LLM-driven algorithm discovery navigates the open-ended and complex search space of algorithms. These findings provide insights to guide the future design of adaptive LLM-driven algorithm generators.
Paper Nr: 136
Title:

Using an Integer Condensed Population for Resource-Constrained Evolution
Authors:

Gary B. Parker, Jay B. Nash and Jim O'Connor
Abstract: Genetic Algorithms (GA) have proven to be versatile tools for solving optimization problems in various domains. However, traditional reliance on fixed-size populations can impose constraints on storage, transmission, and adaptability, particularly in resource‐limited or distributed systems. This paper introduces the Compressed Population Genetic Algorithm (CPGA), a novel approach that replaces traditional population storage with an integer‐array representation, enabling efficient compression and reconstruction of populations. By aggregating the population into a single integer array of bit‐counts rather than a floating‐point probability vector, the CPGA retains exact population statistics while drastically reducing its memory footprint. Compared to probabilistic model–based methods such as the Compact Genetic Algorithm (cGA) or Estimation of Distribution Algorithms (EDA), which require full floating‐point probability distributions and often complex parameter updates, our integer‐array scheme leverages simple, exact count updates. Through controlled modification and regeneration of individuals, the CPGA demonstrates improved scalability and adaptability, especially in distributed and embodied evolutionary systems. Benchmark results on the BBOB‐17 suite show the CPGA matches or exceeds the performance of traditional GAs while offering significant gains in storage efficiency and communication overhead; experiments in the Atari ALE framework further validate its efficacy across disparate domains.
Paper Nr: 146
Title:

The Traveling Tournament Problem: Valid Solutions are Different Across Instance Sizes
Authors:

Bas Loyen, Duncan Bart, Florian Richoux and Daan van den Berg
Abstract: Abstract. We generate all 160 valid solutions for the traveling tournament problem with 4 teams, and randomly select 10,000 valid solutions for 6, 8 and 10 teams. For every number of teams, the difference between any two valid solutions is taken an charted. It turns out these differences are very large, even when a key constraint, the home/away assignment, is completely ignored. These results might signal that the use of metaheuristic algorithms for this problem might be highly problematic.

Area 2 - Applications

Full Papers
Paper Nr: 45
Title:

Evaluating Fitness Averaging Strategies in Cooperative NeuroCoEvolution for Automated Soft Actuator Design
Authors:

Hugo Alcaraz Herrera, Michail-Antisthenis Tsompanas, Igor Balaz and Andrew Adamatzky
Abstract: Soft robotics are increasingly favoured in specific applications such as healthcare, due to their adaptability, which stems from the non-linear properties of their building materials. However, these properties also pose significant challenges in designing the morphologies and controllers of soft robots. The relatively short history of this field has not yet produced sufficient knowledge to consistently derive optimal solutions. Consequently, an automated process for the design of soft robot morphologies can be extremely helpful. This study focusses on the cooperative NeuroCoEvolution of networks that are indirect representations of soft robot actuators. Both the morphologies and controllers represented by Compositional Pattern Producing Networks are evolved using the wellestablished method NeuroEvolution of Augmented Topologies (CPPN-NEAT). The CoEvolution of controllers and morphologies is implemented using the top n individuals from the cooperating population, with various averaging methods tested to determine the fitness of the evaluated individuals. The test-case application for this research is the optimisation of a soft actuator for a drug delivery system. The primary metric used is the maximum displacement of one end of the actuator in a specified direction. Additionally, the robustness of the evolved morphologies is assessed against a range of randomly generated controllers to simulate potential noise in real-world applications. The results of this investigation indicate that CPPN-NEAT produces superior morphologies compared to previously published results from multi-objective optimisation, with reduced computational effort and time. Moreover, the best configuration is found to be CoEvolution with the two best individuals from the cooperative population and the averaging of their fitness using the weighted mean method.
Paper Nr: 169
Title:

Optimization Is not Enough: Why Problem Formulation Deserves Equal Attention
Authors:

Iván Olarte Rodríguez, Gokhan Serhat, Mariusz Bujny, Fabian Duddeck, Thomas Bäck and Elena Raponi
Abstract: Black-box optimization is increasingly used in engineering design problems where simulation-based evaluations are costly and gradients are unavailable. In this context, the optimization community has largely analyzed algorithm performance in context-free setups, while not enough attention has been devoted in how problem formulation and domain knowledge may affect the optimization outcomes. We address this gap through a case study in the topology optimization of laminated composite structures, formulated as a black-box optimization problem. Specifically, we consider the design of a cantilever beam under a volume constraint, with the objective of minimizing compliance while optimizing both the structural topology and fiber orientations. To assess the impact of problem formulation, we explicitly separate topology and material design variables and compare two strategies: a concurrent approach that optimizes all variables simultaneously without leveraging physical insight, and a sequential approach that optimizes variables of the same nature in stages. Our results show that, despite using state-of-the-art optimizers, context-agnostic strategies consistently lead to suboptimal or non-physical designs. In contrast, the sequential strategy yields better-performing and more interpretable solutions. These findings underscore the value of incorporating, when available, domain knowledge into the optimization process and motivate the development of new black-box benchmarks that reward physically informed and context-aware optimization strategies.
Short Papers
Paper Nr: 55
Title:

Edge Weight Learning Approaches for the Promotion of Fairness in the Dictator Game
Authors:

Evan O'Riordan, Colm O'Riordan and Frank G. Glavin
Abstract: In the Ultimatum Game, the proposer issues an ultimatum as to how they wish to split a prize with the responder. The responder either accepts or rejects the proposal. If it is rejected, both players leave empty-handed. The Dictator Game is a variant of the Ultimatum Game where the responder (the recipient) is not given this choice. Instead, they must always accept the proposal from the proposer (the dictator). These games highlight a tension between selfishness and fairness. In this work, we investigate factors that may help induce the evolutionary emergence of fairness in a simulated Dictator Game environment. We employ an Edge Weight Learning mechanism where the weights of the edges connecting the players determine the probability of interactions occurring. We compare multiple approaches that either account for proposal values or for utility. Each approach has two parts: (1) a logical condition to determine whether the edge weight will increase, decrease or remain the same, and (2) a formula to calculate the magnitude of change. Our experimental results show that proposal values are a better stimulus to incorporate into Edge Weight Learning than utility when it comes to romoting fairness.
Paper Nr: 73
Title:

Evolving Neural Controllers for Xpilot-AI Racing Using Neuroevolution of Augmenting Topologies
Authors:

Jim O'Connor, Nicholas Lorentzen, Gary B. Parker and Derin Gezgin
Abstract: This paper investigates the development of high-performance racing controllers for a newly implemented racing mode within the Xpilot-AI platform, utilizing the Neuro Evolution of Augmenting Topologies (NEAT) algorithm. By leveraging NEAT’s capability to evolve both the structure and weights of neural networks, we develop adaptive controllers that can navigate complex circuits under the challenging space simulation physics of Xpilot-AI, which includes elements such as inertia, friction, and gravity. The racing mode we introduce supports flexible circuit designs and allows for the evaluation of multiple agents in parallel, enabling efficient controller optimization across generations. Experimental results demonstrate that our evolved controllers achieve up to 32% improvement in lap time compared to the controller’s initial performance and develop effective racing strategies, such as optimal cornering and speed modulation, comparable to human-like techniques. This work illustrates NEAT’s effectiveness in producing robust control strategies within demanding game environments and highlights Xpilot-AI’s potential as a rigorous testbed for competitive AI controller evolution.
Paper Nr: 111
Title:

Incremental Evolution of Fault-Tolerant Gaits in Octopod Robots
Authors:

Manan B. M. Isak, Gary Parker and Jim O'Connor
Abstract: In this paper, we explore the capability of Genetic Algorithms (GA) to evolve walking gaits for an eight-legged robot with three degrees of freedom per leg when one or more of its legs are disabled. This work is an extension of previous research completed by Parker, Isak and O’Connor which found success in combining GAs and Incremental Evolution to evolve near-optimal gaits for an arachnid-inspired robot. They completed the learning in two increments, where in the first increment individual leg cycles were learned and the second increment facilitated the learning of overall gait coordination. The first increment of our work is the same as in the previous work. We use a Cyclic Genetic Algorithm (CGA) to first learn pulse instructions for the servo motors on each leg to generate leg cycles. In the second increment, which we have improved upon, we use a GA to select the best combination of leg cycles for each leg, learn the timing to execute each leg cycle, and coordinate all of the leg cycles into a single gait. Our improvements in the second increment include adjusting the fitness function and robot model to better reflect how faults affect performance. Training involved disabling up to two legs on the robot in simulation and learning a gait that adapts to these faults. We compare these fault-tolerant gaits to a near-optimal fully-functioning gait. Testing in simulation confirmed the viability of this method.
Paper Nr: 53
Title:

GADA: An Adaptive Genetic Algorithm-Based Framework for Dynamic University Course Timetabling
Authors:

Minh Tam Nguyen, Phat T. Tran-Truong and Anh Truong
Abstract: The University Course Timetabling Problem (UCTP) is a well-known NP-hard multi-objective optimization problem influenced by variety of factors ranging from institutional policies and facilities, institutional facilities, course characteristics to instructor availability, instructor preferences and the diversity and variability of student registration behaviors. This paper introduces GADA, a novel approach to automate course timetabling. Unlike traditional models that mostly assume static, centralized scheduling, GADA is an adaptive genetic algorithm-based approach tailored for decentralized, credit-based systems where students independently register for courses and instructors have diverse time preferences. GADA focuses on optimizing course-to-timeslot allocation while satisfying both hard institutional constraints and soft instructor preferences. By automating critical steps in the scheduling workflow, GADA significantly reduces manual effort, increases scheduling flexibility, and adapts efficiently to late-stage registration changes. It has been implemented and evaluated extensively in the real-world environment at the Faculty of Computer Science and Engineering at Ho Chi Minh City University of Technology (CSE@HCMUT). The experimental results demonstrate GADA’s effectiveness and practical applicability in generating conflict-free and operationally feasible schedules while addressing the complex constraints inherent in decentralized academic timetabling.