Algorithms

22 pages, 530 KB

Open AccessArticle

MEMTIER: Tiered Retrieval, Session-Level Injection, and Typed Consolidation for Long-Running LLM Agents

by Bronislav Sidik and Lior Rokach

Algorithms 2026, 19(7), 607; https://doi.org/10.3390/a19070607 - 22 Jul 2026

Viewed by 314

Long-running large language model (LLM) agents accumulate memory across many sessions, yet most agent runtimes store it as flat text. We present MEMTIER, a tiered memory architecture and consolidation framework for an open-source agent runtime and study three questions: what to store, [...] Read more.

Long-running large language model (LLM) agents accumulate memory across many sessions, yet most agent runtimes store it as flat text. We present MEMTIER, a tiered memory architecture and consolidation framework for an open-source agent runtime and study three questions: what to store, what to inject, and what to keep. First, a tiered episodic–semantic memory reaches an accuracy of 0.382 and an F1 of 0.412 on LongMemEval-S (N = 500) with a 7B model on a 6 GB GPU—a +33-point gain over no retrieval. A controlled single-pipeline ablation shows the gains come from two components—semantic pre-population and two-stage scoping—while the multi-signal retrieval scaffold is, in this setting, an operational pass-through of BM25 top-k; removing it entirely leaves accuracy unchanged. An oracle analysis explains why the system is retrieval-limited: on the diagnostic subset, the correct session is present in the candidate set 98% of the time, so the primary binding constraint is which evidence reaches the reader. A larger generator helps but does not remove the retrieval bottleneck. Second, the bottleneck is injection granularity, not session recall (90.9% session coverage versus 4.5% fact recall@2). Injecting all facts from the top-k retrieved sessions improves multi-session accuracy by +0.120 and knowledge update accuracy by +0.205, and a controlled comparison isolates a structure effect distinct from token quantity. Third, we cast memory consolidation as a Markov decision process, characterize three structural failure modes that render a learned policy uninformative on single-subject benchmarks, and show preliminary evidence on a live agent benchmark that a typed, keyword-based policy recovers near-oracle performance for one memory type (threat pattern memory) without supervision. Type-dependent retention for other memory types is proposed but not yet evaluated. We frame these as related diagnostic studies rather than a single validated pipeline, and, together, they cast agent memory as a pattern recognition problem: recognizing which session patterns carry evidence and which knowledge types to retain. Full article

(This article belongs to the Special Issue Machine Learning for Pattern Recognition (4th Edition))

► Show Figures

Figure 1

18 pages, 683 KB

Open AccessArticle

Dependence-Aware Lag-Resolved Correlation Analysis in Multi-Sensor Stochastic Systems

by Andrew Graham Ryan

Algorithms 2026, 19(7), 606; https://doi.org/10.3390/a19070606 - 22 Jul 2026

Viewed by 313

Abstract

Exploratory lag-resolved correlation analysis is widely used when potential dependence between concurrently recorded signals may be delayed, transient, or weak. In autocorrelated time series, however, scanning across candidate lags creates a multiple-comparison problem, while temporal dependence invalidates naive permutation or parametric correlation tests. [...] Read more.

Exploratory lag-resolved correlation analysis is widely used when potential dependence between concurrently recorded signals may be delayed, transient, or weak. In autocorrelated time series, however, scanning across candidate lags creates a multiple-comparison problem, while temporal dependence invalidates naive permutation or parametric correlation tests. This paper presents a practical dependence-aware scan-level inference procedure for exploratory lag-resolved correlation analysis under temporal autocorrelation. The procedure combines baseline standardisation, lag-resolved Pearson correlation, dependence-preserving surrogate construction (here implemented using block permutation), and max-statistic correction so that inference is performed on the largest absolute correlation observed across the scanned lag domain rather than on post hoc selected lags. The contribution is integrative rather than metric-driven: established components are assembled into a fixed, auditable inference pipeline that preserves within-channel temporal structure while controlling familywise error across exploratory lag scans. The procedure is intended for multi-sensor stochastic systems in which weak synchrony must be distinguished from artefacts of temporal dependence and analytical flexibility. Synthetic null simulations illustrate false-positive inflation under naive lag scanning, characterise calibration through a block length sensitivity analysis, and compare complete inference pipelines, demonstrating that, under the dependence regimes examined here, valid exploratory lag inference requires scan-level multiplicity control in addition to dependence-preserving surrogate generation. Full article

► Show Figures

Figure 1

18 pages, 3494 KB

Open AccessArticle

Towards Rotated Object Detection with Pose-Aware and Dense Feature Modulation

by Dehua Bai, Donglin Jing and Meng Zhao

Algorithms 2026, 19(7), 605; https://doi.org/10.3390/a19070605 - 22 Jul 2026

Viewed by 296

Abstract

Remote sensing image object detection is a core task in computer vision, which plays a vital role in intelligent transportation, port monitoring, and infrastructure management. However, rotated dense objects in remote sensing scenes suffer from severe challenges, including arbitrary 0–360° pose variations, large-scale [...] Read more.

Remote sensing image object detection is a core task in computer vision, which plays a vital role in intelligent transportation, port monitoring, and infrastructure management. However, rotated dense objects in remote sensing scenes suffer from severe challenges, including arbitrary 0–360° pose variations, large-scale differences, dense spatial aggregation, and blurred boundaries. Traditional Convolutional Neural Networks rely on fixed sampling grids and receptive fields, failing to adaptively capture the dynamic morphological and pose features of tilted targets. Meanwhile, existing methods struggle to address feature coupling and boundary misjudgment among densely arranged objects, leading to degraded detection accuracy. To tackle these bottlenecks, we propose an adaptive detection framework named PDNet for rotated dense object detection. The framework integrates four key designs: First, a Pose-Aware Dynamic Sampling Mechanism (PDSM) is developed to estimate the target principal axis in real time and learn a deformable offset field, which dynamically adjusts the convolution sampling pattern and receptive field shape to adapt to target pose variations. Second, a Dense-Scene Feature Modulation Mechanism (DSFM) constructs a dynamic weight field based on local feature responses to enhance discriminative target features and suppress inter-target interference in dense regions. Third, the Strip-Based Context Attention (SCA) module fuses global and local contextual information to strengthen the representation of small and weak targets. Fourth, a boundary-aware rotation loss function is designed to optimize the regression accuracy of rotated bounding boxes via pixel-level supervision. Extensive experiments on DOTA-v1.0, AI-TOD achieve 79.86% mAP and 25.95% AP, outperforming state-of-the-art methods. Full article

(This article belongs to the Special Issue Advances in Deep Learning-Based Data Analysis)

► Show Figures

Figure 1

38 pages, 3935 KB

Open AccessArticle

Physics-Informed Virtual Sensing for Wind Turbine Blade Degradation Detection Using Flapwise–Edgewise Load Coupling

by Attia Bibi, Chiheng Huang, Wenxian Yang and Rakesh Mishra

Algorithms 2026, 19(7), 604; https://doi.org/10.3390/a19070604 - 21 Jul 2026

Viewed by 216

Abstract

Early detection of blade structural degradation is critical for wind turbine reliability, as unplanned failures contribute to approximately 25% of operational downtime. Traditional blade condition monitoring methods based on distributed vibration, strain, or acoustic emission sensing often suffer from high hardware complexity and [...] Read more.

Early detection of blade structural degradation is critical for wind turbine reliability, as unplanned failures contribute to approximately 25% of operational downtime. Traditional blade condition monitoring methods based on distributed vibration, strain, or acoustic emission sensing often suffer from high hardware complexity and substantial data-processing requirements. Consequently, blade load measurements at the blade root have emerged as a promising and more practical alternative for structural health monitoring. This study introduces a physics-informed virtual sensing framework that combines high-fidelity aeroelastic simulation with machine learning trained exclusively on healthy operational data for blade degradation detection. The proposed method is based on the premise that, under healthy conditions, the relationship between flapwise and edgewise bending moments remains consistent across operating conditions, whereas structural degradation disrupts this coupling and induces measurable deviations from the learned correlation. To implement this concept, OpenFAST simulations of the NREL 5MW reference turbine are conducted under multiple operating conditions with progressively introduced stiffness degradation, forming a comprehensive dataset. A Random Forest model is trained to predict edgewise bending moments from flapwise loads and structural parameters using only healthy-condition data, achieving high predictive accuracy under healthy operating conditions (R² = 0.98; MAPE = 3.67%). Fault detection is performed by analysing residuals between predicted and simulated edgewise bending moments, which increase systematically when degradation is present. The proposed framework achieved an overall classification accuracy of 88.9%, sensitivity of 88.9%, precision of 96.0%, F1-score of 0.923, and an AUC of 0.951 across all degradation levels and operating conditions, while achieving 100% detection of severe (20%) torsional stiffness degradation cases. Notably, degradation reduces absolute blade loads due to aeroelastic twist-to-feather coupling, yet introduces distinctive load asymmetry patterns that enable effective discrimination. The method demonstrates strong generalisation across the investigated wind speed conditions without retraining. By reconstructing unmeasured edgewise loads from readily available flapwise measurements, the proposed approach enables scalable structural health monitoring of existing wind turbine fleets without hardware retrofitting, offering a practical pathway toward physics-guided digital twin development. Full article

(This article belongs to the Special Issue AI-Powered Predictive Maintenance: Transforming Industrial Operations Through Intelligent Fault Diagnosis (2nd Edition))

► Show Figures

Figure 1

45 pages, 5047 KB

Open AccessArticle

TRT-GLA: Tri-Representation Transformers with Global–Local Attention for High-Fidelity Multi-Modal MRI Super-Resolution

by Suhaila Abuowaida, Hamza Abu Owida, Tareq Hamadneh, Nawaf Alshdaifat, Hamza A. Mashagba, Mwaffaq Abu Alhaija and Azlan B. Abd Aziz

Algorithms 2026, 19(7), 603; https://doi.org/10.3390/a19070603 - 21 Jul 2026

Viewed by 254

Abstract

The super-resolution (SR) of Magnetic Resonance Imaging (MRI) is essential for utilizing clinical scans with limited resolution, noise, and anisotropic sampling, such as multi-modal brain tumor imaging. In this work, we propose a Tri-Representation hybrid framework for MRI SR, TRT-GLA, that redefines the [...] Read more.

The super-resolution (SR) of Magnetic Resonance Imaging (MRI) is essential for utilizing clinical scans with limited resolution, noise, and anisotropic sampling, such as multi-modal brain tumor imaging. In this work, we propose a Tri-Representation hybrid framework for MRI SR, TRT-GLA, that redefines the MRI SR task as a joint spatial–spectral–structural high-resolution image generation problem. TRT-GLA utilizes (i) spatial global–local attentions for modeling the spatial anatomy, (ii) a Fourier spectral transfer mechanism for upholding spectral consistency, and (iii) multi-scale hierarchical spectral decomposition for improved edge details. To adapt the learning framework to medical imaging characteristics, we introduce a tri-representation consistent loss function that explicitly combines pixel-wise, spectral, and edge structure priors from the high-resolution ground-truth, as well as a progressive resolution learning strategy. Our large-scale brain tumor experiments, on the IXI, BraTS 2019, 2020, and 2023 datasets, show that TRT-GLA achieves state-of-the-art results at upsampling factors of ×2, ×4, and ×8, respectively, achieving substantial improvements across CNN, GAN, and transformer-based methods in Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Multi-scale Structural Similarity Index (MS-SSIM). We further demonstrate how SR benefits brain tumor segmentation through the downstream task evaluation of a dual-branch segmentation framework. TRT-GLA produces highly accurate tumor segmentation results from low-resolution inputs, improving over native high-resolution inputs at ×8 in critical tumor boundary regions and in small tumor regions. There remains a small gap between native, high-resolution imaging and SR-enhanced performance, which TRT-GLA nearly closes under realistic scenarios. Our results highlight the importance of synthesizing unified priors over spatial, spectral, and structural domains within a transformer for anatomically faithful reconstructions. Importantly, we also establish the utility of TRT-GLA in supporting quantitative analysis through a downstream tumor segmentation experiment that is clinically relevant. Full article

(This article belongs to the Special Issue Artificial Intelligence in Sustainable Development)

► Show Figures

Figure 1

15 pages, 3719 KB

Open AccessArticle

Exploratory Study of Triple-Entry Data (X-STATIS) for the Evaluation of Energy Quality in Induction Motors Under Simulated Faults

by Alexandra Gabriela Valenzuela-Cobos, Jorge Fabricio Guevara-Viejó, Delia Noriega Verdugo, Juan Diego Valenzuela-Cobos, Purificación Vicente-Galindo and Purificación Galindo-Villardón

Algorithms 2026, 19(7), 602; https://doi.org/10.3390/a19070602 - 21 Jul 2026

Viewed by 248

Abstract

Energy, conceived as an essential element for economic and social progress, underpins the manufacture of products, transport, the provision of basic needs and, consequently, the improvement of society’s well-being. Within this framework, three-phase induction motors (MIT) play a key role in industrial processes. [...] Read more.

Energy, conceived as an essential element for economic and social progress, underpins the manufacture of products, transport, the provision of basic needs and, consequently, the improvement of society’s well-being. Within this framework, three-phase induction motors (MIT) play a key role in industrial processes. However, these devices can suffer from various types of faults that negatively affect their performance and durability, including deterioration of the stator winding and obstruction of the cooling system. This research focuses on the analysis of various operating conditions of induction motors through the application of the X-STATIS statistical technique, with the aim of determining the capacity of electrical and mechanical indicators to detect faults. To this end, eight experimental situations were considered: one corresponding to normal operating conditions and seven associated with induced faults, evaluated under six different voltage levels. The variables recorded included three-phase current, efficiency, torque, and speed, organized into a data cube consisting of six tables, one for each voltage level, analyzed in the RStudio environment. The study of the interstructure allowed us to identify two main behavioral trends: a set of faults consisting of ROT2, INS4 and INS5, and another consisting of NON1, MECH6, MECH7, VENT8 and STA3. This classification was validated by high RV coefficient values, above 0.91, which shows high structural consistency between the different simulated scenarios. In the engagement stage, significant positive correlations were observed between phase currents (CF1, CF2 and CF3) and torque (TOR), as well as an inverse relationship between speed (SPE) and torque. Efficiency (EFF), on the other hand, showed a different behavior, with a greater association with high voltage levels (V4 and V5), highlighting its sensitivity to conditions close to the nominal load. Finally, the analysis of the internal structure allowed the projection of the different failure conditions within the space defined by the compromise to be examined. Full article

(This article belongs to the Special Issue Algorithms in Data Classification (3rd Edition))

► Show Figures

Figure 1

25 pages, 649 KB

Open AccessArticle

A Computational Framework to Assess Model Complexity Trade-Offs in Country-Level Temperature Anomaly Time Series

by Rafael Rojas-Galván, Luis E. Gallo-Gonzalez, Juan S. Arteaga-Hernandez, Omar Rodríguez-Abreo and Juvenal Rodríguez-Reséndiz

Algorithms 2026, 19(7), 601; https://doi.org/10.3390/a19070601 - 20 Jul 2026

Viewed by 274

Abstract

Accurate forecasting of country-level temperature anomalies is increasingly important for climate monitoring, policy planning, and environmental risk assessment. However, the trade-off between predictive performance, model complexity, and computational cost remains insufficiently explored, particularly across multiple countries using compact and interpretable feature representations. This [...] Read more.

Accurate forecasting of country-level temperature anomalies is increasingly important for climate monitoring, policy planning, and environmental risk assessment. However, the trade-off between predictive performance, model complexity, and computational cost remains insufficiently explored, particularly across multiple countries using compact and interpretable feature representations. This study presents a comprehensive comparative evaluation of eight forecasting approaches for annual temperature anomaly prediction using country-level observations from the FAOSTAT Temperature Change dataset. The evaluated methods comprise a Persistence baseline, Ordinary Least Squares (OLS), Ridge regression, Support Vector Regression (SVR), Random Forest, a multilayer perceptron (MLP), and the classical time-series models ARIMA and ETS. Annual temperature anomalies were modeled using lagged observations, a temporal trend, and a trailing moving average under a temporally ordered 80/20 train–test split. Model performance was assessed using RMSE, MAE,

R^{2}

, per-country win-rate, computational runtime, and pairwise statistical comparisons based on the Wilcoxon signed-rank test with Holm correction. Hyperparameters were optimized through expanding-window temporal cross-validation, and an ablation study was conducted to quantify feature contributions. Results indicate that the ETS model achieved the best overall predictive performance, obtaining the lowest median RMSE (0.3388 °C), the lowest MAE (0.2792 °C), and the highest per-country win-rate (40.07%). ARIMA provided competitive forecasting accuracy but incurred substantially higher computational cost, whereas OLS and Ridge offered an attractive compromise between predictive performance, robustness, interpretability, and computational efficiency. In contrast, the more flexible machine learning models (SVR, Random Forest, and MLP) did not consistently outperform the simpler approaches despite their higher complexity. Overall, the results demonstrate that classical statistical forecasting methods remain highly competitive for annual country-level temperature anomaly prediction and that increasing model complexity does not necessarily translate into improved predictive performance. Full article

(This article belongs to the Special Issue Artificial Intelligence Algorithms in Sustainability)

► Show Figures

Figure 1

22 pages, 2732 KB

Open AccessArticle

L21RSVD: Robust L21 Norm SVD-Type Latent Factor Models for Rating Prediction

by Chenggang He, Can Hu and Demeng Qian

Algorithms 2026, 19(7), 600; https://doi.org/10.3390/a19070600 - 20 Jul 2026

Viewed by 184

Abstract

Recommender systems face fundamental challenges, including extreme data sparsity and noisy rating observations. We propose L21RSVD, a robust latent factor model that employs L21 norm regularization within a Singular Value Decomposition framework to mitigate the impact of outliers while preserving low-rank structure. Unlike [...] Read more.

Recommender systems face fundamental challenges, including extreme data sparsity and noisy rating observations. We propose L21RSVD, a robust latent factor model that employs L21 norm regularization within a Singular Value Decomposition framework to mitigate the impact of outliers while preserving low-rank structure. Unlike conventional L2-regularized approaches, our formulation induces group sparsity in the latent factor space, yielding more discriminative user and item representations. We derive three optimization variants: standard L21RSVD, L21RSVD without the squared term, and coefficient-free adaptive L21RSVD. Building upon these, we introduce a fusion strategy that adaptively aggregates predictions based on local data density. Extensive experiments on benchmark datasets demonstrate that L21RSVD substantially outperforms classical collaborative filtering and the SVD-type model. The proposed fusion model achieves state-of-the-art performance, reducing RMSE by up to 31.21% and MAE by up to 42.99% relative to baseline methods. Full article

► Show Figures

Figure 1

25 pages, 3847 KB

Open AccessArticle

Temporal Super-Resolution of Non-Stationary Signals: Mixed-Domain Training and a Hybrid Wavelet–Superlet Pilot

by Julio Ibarra-Fiallo, D’hamar Agudelo-Moreno and Juan A. Lara

Algorithms 2026, 19(7), 599; https://doi.org/10.3390/a19070599 - 20 Jul 2026

Viewed by 259

Abstract

Temporal super-resolution (SR) aims to reconstruct a high-resolution signal from a low-resolution observation. When hardware limits force low sampling rates, this problem becomes critical for non-stationary signals with abrupt transients and rapid spectral changes. This manuscript reports a deterministic case study using pretrained [...] Read more.

Temporal super-resolution (SR) aims to reconstruct a high-resolution signal from a low-resolution observation. When hardware limits force low sampling rates, this problem becomes critical for non-stationary signals with abrupt transients and rapid spectral changes. This manuscript reports a deterministic case study using pretrained 1D convolutional models and deterministic evaluation on paired real, synthetic, and mixed EEG-like signals. A compact encoder–linear upsampler–refinement architecture is evaluated at 5× upsampling under four training regimes: synthetic-only, real-only, tuned-real, and mixed. Performance is assessed with Mean Squared Error (MSE), Mean Absolute Error (MAE), Normalized Mean Absolute Error (NMAE), Log Spectral Distance (LSD), and spectral correlation (SCORR). Across 12 model–dataset combinations, mixed-domain training yields the most robust cross-domain behavior, outperforming single-domain checkpoints on real and mixed evaluation subsets. These findings support the practical value of training corpus composition for temporal SR under distribution shift. A focused morphological event analysis further shows that reconstruction error concentrates at abrupt amplitude and frequency boundaries, confirming that these transient regions are the dominant local challenge. An exploratory hybrid wavelet–superlet pilot is also reported; it achieves competitive pointwise error on selected domains but exhibits a substantial spectral-fidelity gap, indicating that frequency-aware inputs alone do not guarantee spectral reconstruction without auxiliary spectral losses. Full article

(This article belongs to the Special Issue Machine Learning Algorithms for Signal Processing)

► Show Figures

Figure 1

23 pages, 6226 KB

Open AccessArticle

Generalization-Enhanced State Assessment of Railway Power Transformers Using Feature-Guided Stacking Learning

by Yuanfang Huang, Zhanhong Huang and Junbin Chen

Algorithms 2026, 19(7), 598; https://doi.org/10.3390/a19070598 - 20 Jul 2026

Viewed by 242

Abstract

Reliable state assessment of railway traction power transformers is challenged by heterogeneous operating environments, measurement disturbances, coupled gas-generation mechanisms, and uneven fault-sample distributions. Conventional dissolved gas analysis (DGA) ratio rules and single-model classifiers often show insufficient generalization when rare faults and boundary-ambiguous operating [...] Read more.

Reliable state assessment of railway traction power transformers is challenged by heterogeneous operating environments, measurement disturbances, coupled gas-generation mechanisms, and uneven fault-sample distributions. Conventional dissolved gas analysis (DGA) ratio rules and single-model classifiers often show insufficient generalization when rare faults and boundary-ambiguous operating states are encountered. To address this issue, this paper proposes a feature-guided stacking framework for state assessment of oil-immersed railway power transformers. First, a DGA-oriented fusion-feature representation is established by combining raw gas concentrations, gas-ratio descriptors, and an aggregated dissolved-gas analysis factor. Second, DBSCAN-assisted sample structuring is introduced to identify density patterns, sparse rare fault regions, and boundary samples, thereby improving the organization of imbalanced monitoring records. Third, a monitoring-feature-embedded stacking model is developed in which heterogeneous base learners are adaptively weighted according to feature-reliability information and integrated through a cross-validated meta-learner. This synthetic-data-based validation provides a controlled and reproducible proof-of-concept. Therefore, the reported results should be interpreted as evidence of methodological feasibility. Under the default synthetic setting, the proposed feature-guided stacking (FE-stacking) method achieves an accuracy of 99.70% and a macro-F1 of 99.55%. Under the severe minority-retention setting in which only 25% of low-energy discharge (LD) and low-temperature overheating (LT) training samples are preserved, it obtains an accuracy of 99.62%, a macro-F1 of 99.40%, and an LT recall of 96.61%, slightly surpassing random forest (RF) and outperforming Original Stacking in rare fault robustness. These results indicate that feature-guided ensemble learning can improve the generalization stability of DGA-based transformer state assessment under imbalanced and boundary-ambiguous conditions. From a practical perspective, the proposed framework can serve as a decision-support module for transformer condition screening, maintenance prioritization, and alarm verification in railway traction power-supply systems. Full article

► Show Figures

Figure 1

31 pages, 16749 KB

Open AccessArticle

A Deep Tabular Learning Framework for E-Commerce Delivery Delay Prediction and Cost-Sensitive Operational Risk Mitigation

by Yuxuan Zhang, Yangxin Wang and Yuanyuan Wang

Algorithms 2026, 19(7), 597; https://doi.org/10.3390/a19070597 - 19 Jul 2026

Viewed by 310

Abstract

Mitigating delayed delivery risks is a critical imperative for e-commerce platforms to prevent financial losses and optimize customer lifetime value. However, extracting actionable intelligence from heterogeneous tabular logistics data remains a persistent challenge. Therefore, the specific objective of this research is to develop [...] Read more.

Mitigating delayed delivery risks is a critical imperative for e-commerce platforms to prevent financial losses and optimize customer lifetime value. However, extracting actionable intelligence from heterogeneous tabular logistics data remains a persistent challenge. Therefore, the specific objective of this research is to develop a deep tabular learning framework capable of accurately predicting last-mile delivery delays and translating these predictions into cost-sensitive operational intervention strategies. To achieve this, we propose the Gated Feature Interaction Multilayer Perceptron (GFI-MLP), which leverages dual-branch gated interactions and hidden-channel recalibration to learn complex fulfillment risk representations. Validated on a public e-commerce logistics dataset (Kaggle), GFI-MLP achieves an AUC of 0.8464 and an F1-score of 0.8236, significantly outperforming eight baseline models (including Gradient-Boosting and Tabular Transformer architectures). Bridging algorithmic innovation with operational management, we introduce a dynamic cost-benefit simulation to translate predictive probabilities into quantifiable economic value. Ultimately, GFI-MLP provides a robust, data-driven tool to minimize proactive intervention costs and mitigate operational risks. Full article

► Show Figures

Figure 1

23 pages, 912 KB

Open AccessArticle

Acceptance Rate-Adaptive pCN MCMC for Moderate-Dimensional Bayesian Inverse Problems

by Shucan Xia and Haoran Song

Algorithms 2026, 19(7), 596; https://doi.org/10.3390/a19070596 - 19 Jul 2026

Viewed by 195

Abstract

The preconditioned Crank–Nicolson (pCN) Markov chain Monte Carlo method is a standard tool for high-dimensional Bayesian inverse problems because its proposals preserve the prior and its performance degrades gracefully under mesh refinement. Yet its efficiency depends heavily on the step-size parameter

β

, [...] Read more.

The preconditioned Crank–Nicolson (pCN) Markov chain Monte Carlo method is a standard tool for high-dimensional Bayesian inverse problems because its proposals preserve the prior and its performance degrades gracefully under mesh refinement. Yet its efficiency depends heavily on the step-size parameter

β

, which practitioners must tune manually for each problem. To remove this burden, we propose an acceptance rate-adaptive pCN algorithm that calibrates

β

automatically during a finite burn-in phase via a Robbins–Monro stochastic approximation driven by cumulative acceptance rate feedback. We also studied a dual-signal extension that supplements acceptance rate adaptation with an auxiliary effective sample size (ESS) feedback term, activated gradually through a smooth logistic switch. To contextualize the proposed methods, we conducted a systematic comparison with the adaptive Metropolis (AM) and the Metropolis-adjusted Langevin algorithm (MALA). In both adaptive pCN variants, adaptation is confined to the burn-in and the step size is then frozen, so that the post-burn-in samples are produced by a time-homogeneous Metropolis–Hastings kernel targeting the correct posterior. Numerical experiments on five linear-Gaussian benchmarks with a controlled spectral structure (dimensions: 10 to 100, condition numbers: 2 to 200, SNR from

- 9

dB to

+ 8

dB) and a nonlinear PDE-constrained inverse problem governed by the cubic–quintic nonlinear Schrödinger equation showed that the acceptance rate-adaptive pCN yields consistent improvements over vanilla pCN, reaching approximately an

1.8 \times

higher ESS in ill-conditioned regimes with low inter-seed variance. The AM achieved a superior ESS on well-conditioned and low-dimensional problems, while the MALA excelled in low-SNR regimes, but degraded under sharp posteriors. A sensitivity analysis confirmed the robustness to hyperparameter choices, and multi-seed experiments validated the reproducibility of all findings. These results indicate that automatic step-size adaptation can substantially improve the practical efficiency of pCN for challenging moderate-dimensional Bayesian inverse problems without manual tuning. Full article

► Show Figures

Figure 1

22 pages, 1341 KB

Open AccessArticle

Boundary Sliding-Mode Saturated Control for a Flexible Structure System with External Disturbances

by Wei Wu, Qian Ye, Yulin Wu and Xuyang Lou

Algorithms 2026, 19(7), 595; https://doi.org/10.3390/a19070595 - 19 Jul 2026

Viewed by 308

Abstract

This work deals with the boundary stabilization problem of a flexible beam system subject to actuator saturation and external disturbances. Firstly, for the case of actuator input saturation, by introducing a dead-zone function and constructing an auxiliary system, the influence of saturation nonlinearity [...] Read more.

This work deals with the boundary stabilization problem of a flexible beam system subject to actuator saturation and external disturbances. Firstly, for the case of actuator input saturation, by introducing a dead-zone function and constructing an auxiliary system, the influence of saturation nonlinearity is compensated. Secondly, a boundary saturated controller is proposed. Meanwhile, for an actuator-saturated beam system subject to unknown bounded disturbances, the control objective is to attenuate their effects. Then, a sliding-mode saturated controller is constructed based on the boundary saturated control strategy. The uniform bounded stability of the system is proven in the sense of Lyapunov. The simulation results verify the effectiveness of the proposed methods. Full article

(This article belongs to the Section Algorithms for Multidisciplinary Applications)

► Show Figures

Figure 1

25 pages, 7930 KB

Open AccessArticle

From Forecasting Accuracy to Trading Profitability: Evaluating Sequence Models for Stock Price Prediction

by Carol Anne Hargreaves and Hieu Le Trung

Algorithms 2026, 19(7), 594; https://doi.org/10.3390/a19070594 - 18 Jul 2026

Viewed by 316

Abstract

Accurate stock price forecasting remains a challenging problem due to the noisy, nonlinear, and non-stationary characteristics of financial time series. Although recent advances in deep learning have improved predictive capabilities, most prior studies evaluate forecasting models primarily using statistical error metrics, with limited [...] Read more.

Accurate stock price forecasting remains a challenging problem due to the noisy, nonlinear, and non-stationary characteristics of financial time series. Although recent advances in deep learning have improved predictive capabilities, most prior studies evaluate forecasting models primarily using statistical error metrics, with limited consideration of their practical value in trading and investment decision-making. This creates a gap between predictive performance and real-world economic utility. This study proposes a decision-oriented evaluation framework for multi-step stock price forecasting that jointly assesses predictive accuracy and trading profitability within a unified experimental setting. Using data from 91 ASX 100 stocks after data cleaning, with a testing period spanning 2019–2020, several deep learning architectures, including Multi-Layer Perceptron (MLP), Gated Recurrent Unit (GRU), Seq2Seq, and attention-based sequence models, are systematically compared under identical training and trading conditions. The results show that the Seq2Seq model achieved the best overall performance, obtaining the lowest average MAPE of 0.0293 and the highest ROI of 23.2%, while the attention-based model achieved a similar MAPE of 0.0294 but a lower ROI of 12.4%. Although differences in forecasting accuracy were relatively small, the Seq2Seq model achieved the highest observed trading profitability and generated a higher observed return than a passive market benchmark under the proposed evaluation framework. These findings suggest that evaluation based solely on prediction accuracy may not fully capture the practical value of forecasting models. Full article

(This article belongs to the Special Issue AI-Driven Business Analytics Revolution)

► Show Figures

Figure 1

45 pages, 2940 KB

Open AccessArticle

Explorations on Improving Interpretability of Decision Making Processes of Rule-Based Classifiers

by Urszula Stańczyk

Algorithms 2026, 19(7), 593; https://doi.org/10.3390/a19070593 - 17 Jul 2026

Viewed by 199

Abstract

Rule-based classifiers are often preferred over other types of learners due to the transparent mode in which decisions are made. Each decision rule includes in its premise conditions on attributes. When they are satisfied, the conclusion part of the rule comes into play [...] Read more.

Rule-based classifiers are often preferred over other types of learners due to the transparent mode in which decisions are made. Each decision rule includes in its premise conditions on attributes. When they are satisfied, the conclusion part of the rule comes into play and leads to assigning an object to a specific class. Following the classification process is relatively straightforward but can become more complex when the cardinality of rule set is high. Furthermore, when rules are induced from continuous data, the conditions listed belong to this domain as well, which makes them less general. This paper presents an illustrative example for the exploratory research methodology where the sets of rules are induced in the continuous input domain, but next, they are transformed by discretisation procedures, which results in a simplified representation of the data and knowledge patterns learnt. In addition, the rule sets are also filtered based on rankings obtained for variants of the transformed data. The processing results in reduced decision algorithms with categorical conditions. This simplification is advantageous in and of itself, but the experiments carried out on datasets in the stylometric domain show that it can also lead to enhanced performance of rule-based classifiers. Full article

► Show Figures

Figure 1

13 pages, 316 KB

Open AccessArticle

Exploratory Factor Analysis by Gauss-Newton

by Kenneth Lange

Algorithms 2026, 19(7), 592; https://doi.org/10.3390/a19070592 - 17 Jul 2026

Viewed by 186

Abstract

After more than a century, factor analysis remains one of the most popular tools in applied statistics. Exploratory factor analysis tends to be driven by iterated principal axis factorization. Confirmatory factor analysis tends to rely on maximum likelihood estimation. This paper demonstrates the [...] Read more.

After more than a century, factor analysis remains one of the most popular tools in applied statistics. Exploratory factor analysis tends to be driven by iterated principal axis factorization. Confirmatory factor analysis tends to rely on maximum likelihood estimation. This paper demonstrates the computational superiority of the Gauss–Newton method in minimizing the principal axis loss. On sample problems, Gauss–Newton is more than three orders of magnitude faster than popular implementations of maximum likelihood factor analysis. As part of this comparison, we derive an alternative Gauss–Newton method that leverages the MM principle of optimization. We also explore a simple perturbation correction that dramatically improves the accuracy of the factor loading matrices derived from approximate spectral decompositions. Finally, we suggest a robust version of iterated principal axis factorization that leverages the MM principle, block descent, and the Gauss–Newton method. These innovations are implemented in Julia code and applied to a sequence of representative problems. Full article

35 pages, 5016 KB

Open AccessArticle

Algorithms for Smart-City Waste Infrastructure: A Two-Stage Stochastic MILP with Endogenous Waste-to-Energy Sizing and Shadow-Price Policy Design for Metropolitan Athens

by Athanasios S. Dagoumas

Algorithms 2026, 19(7), 591; https://doi.org/10.3390/a19070591 - 17 Jul 2026

Viewed by 208

Abstract

Decarbonising municipal solid waste (MSW) is a defining algorithmic challenge for smart cities: waste-to-energy (WtE), composting and material-recovery investments must be committed years ahead under deep uncertainty about household source-separation uptake and the governing policy instruments (landfill taxes, carbon prices, compost subsidies). We [...] Read more.

Decarbonising municipal solid waste (MSW) is a defining algorithmic challenge for smart cities: waste-to-energy (WtE), composting and material-recovery investments must be committed years ahead under deep uncertainty about household source-separation uptake and the governing policy instruments (landfill taxes, carbon prices, compost subsidies). We present a two-stage stochastic mixed-integer linear programming (MILP) framework, applied to the Attica region of Greece (Athens; 5122 t/day MSW) and calibrated to confirmed 2024 weighbridge data. The contribution is an integration strategy rather than a new technique: endogenous WtE capacity sizing (Special Ordered Sets of Type 2 (SOS2) piecewise-linear cost, economies-of-scale exponent 0.85), the bilinear capacity–build coupling linearised exactly by McCormick envelopes (one factor being binary), and Pigouvian shadow-price recovery of the optimal policy instruments are combined in a single-shot, gap-bounded MILP and embedded in a 10,000-run Latin-hypercube Monte Carlo layer over 12 parameters with Spearman sensitivity indices. The individual components are established; their joint formulation is, to our knowledge, new. The pipeline solves 10,415 MILP instances. Three results are policy-relevant: investment is robust to rollout uncertainty (VSS ≈ €0; EVPI ≈ €5.4 M, 0.15%); the carbon price alone explains ~80% of cost variance (ρ = +0.891); and, under the model’s calibration, the implied Pigouvian-optimal landfill tax (€1100–3300/t) indicates a binding landfill cap is needed to secure diversion. The framework transfers to any metropolitan MSW system facing decarbonisation and circular-economy mandates. Full article

(This article belongs to the Special Issue Algorithms for Smart Cities (3rd Edition))

► Show Figures

Figure 1

38 pages, 6583 KB

Open AccessArticle

An Intelligent Fuzzy Third-Order Sliding Mode Strategy for Energy Management of DFIG-Based Wind Energy Systems

by Abdelaziz Belhait, Messaoud Louafi, Habib Benbouhani, Sihem Ghoudelbourk, Abdessmad Milles, Ghoulemallah Boukhalfa and Nicu Bizon

Algorithms 2026, 19(7), 590; https://doi.org/10.3390/a19070590 - 16 Jul 2026

Viewed by 240

Abstract

The doubly fed induction generator (DFIG) has become one of the most widely adopted technologies in modern wind energy conversion systems due to its high efficiency, flexible operation, and capability to operate under variable wind-speed conditions. Nevertheless, maintaining high power quality, reducing harmonic [...] Read more.

The doubly fed induction generator (DFIG) has become one of the most widely adopted technologies in modern wind energy conversion systems due to its high efficiency, flexible operation, and capability to operate under variable wind-speed conditions. Nevertheless, maintaining high power quality, reducing harmonic distortion, and ensuring reliable operation under parameter uncertainties and grid disturbances remain major challenges. Conventional proportional–integral (PI)-based control approaches often suffer from limited robustness and sensitivity to system nonlinearities, which can adversely affect the dynamic performance and operational stability of wind energy systems. To address these limitations, this paper proposes an advanced hybrid control strategy based on the integration of fuzzy logic (FL) and a third-order sliding mode controller (TOSMC) for the control of DFIG-based wind turbines. The proposed FL–TOSMC combines the robustness and fast convergence properties of sliding mode control with the adaptive capability of fuzzy logic, thereby reducing dependence on accurate mathematical models and enhancing tolerance to parameter variations and external disturbances. The developed controller is applied to maximum power point tracking (MPPT) and direct field-oriented control of the DFIG to maximize energy extraction and ensure stable power regulation. The simulation results obtained in the MATLAB/Simulink (2021) environment demonstrate that the proposed strategy improves the dynamic response of the system by reducing overshoot, accelerating error convergence, minimizing steady-state oscillations, and decreasing total harmonic distortion compared with the conventional TOSMC approach. Furthermore, the proposed controller provides enhanced tracking performance, improved robustness against parameter uncertainties, and better power-quality characteristics under various operating conditions. These results indicate that the FL–TOSMC approach is an effective control solution for improving the operational performance of DFIG-based wind energy conversion systems. Full article

(This article belongs to the Special Issue Advanced Intelligent Control Algorithms for Power, Energy, and Cyber-Physical Systems)

► Show Figures

Figure 1

36 pages, 3023 KB

Open AccessArticle

Application of the Mushroom-Picking Optimization Algorithm to the Optimal Control of Discrete-Time Deterministic and Stochastic Dynamical Systems

by Andrei V. Panteleev and Ivan S. Nadorov

Algorithms 2026, 19(7), 589; https://doi.org/10.3390/a19070589 - 16 Jul 2026

Viewed by 222

Abstract

A new method for finding the global extremum of functions of many variables under interval constraints is proposed. The method simulates the process of foraging for edible mushrooms in a forest by a group of mushroom pickers. The algorithm includes a forest exploration [...] Read more.

A new method for finding the global extremum of functions of many variables under interval constraints is proposed. The method simulates the process of foraging for edible mushrooms in a forest by a group of mushroom pickers. The algorithm includes a forest exploration stage (a set of feasible solutions) to find mushroom sites and an exploitation stage, during which previously identified mushroom sites are intensively explored by implementing various movement strategies for the mushroom pickers. The method is classified as both bio-inspired metaheuristic algorithm and multi-agent evolutionary algorithm. Its effectiveness is demonstrated using eight typical problems of static parametric optimization of technical systems: pressure vessel problem, welded beam optimization problem, tension/compression spring design problem, gear train optimization design, speed reducer problem, three-bar truss problem, tubular column design optimization, flapping wing design optimization. Experience solving these problems allowed us to formulate recommendations for defining the hyperparameters of the proposed optimization method. The primary goal of the development was to apply this new bio-inspired optimization algorithm to three optimal control problems for discrete dynamic systems: optimal control of a single trajectory, a bundle of trajectories of a deterministic system starting from a given set of initial states, and optimal control of stochastic systems. Full article

(This article belongs to the Special Issue Computational Intelligence and Nature Inspired Algorithms)

► Show Figures

Figure 1

36 pages, 12976 KB

Open AccessArticle

Optimization of Low-Channel EEG Configurations and Temporal Segmentation for Motor Imagery Classification Using a Flexible EEGNet Framework

by Yelnur Tuimebay, Chingiz Alimbayev, Zhadyra Alimbayeva and Kassymbek Ozhikenov

Algorithms 2026, 19(7), 588; https://doi.org/10.3390/a19070588 - 16 Jul 2026

Viewed by 305

Abstract

Motor imagery-based brain–computer interfaces (BCIs) have attracted significant attention due to their potential applications in assistive technologies, neurorehabilitation, and wearable human–machine interaction systems. However, practical implementation of EEG-based BCIs remains challenging because high-density EEG recordings increase hardware complexity, prolong setup time, and introduce [...] Read more.

Motor imagery-based brain–computer interfaces (BCIs) have attracted significant attention due to their potential applications in assistive technologies, neurorehabilitation, and wearable human–machine interaction systems. However, practical implementation of EEG-based BCIs remains challenging because high-density EEG recordings increase hardware complexity, prolong setup time, and introduce substantial spatial redundancy. In addition, the performance of motor imagery classification strongly depends on preprocessing strategy and temporal segmentation parameters. Unlike previous studies that primarily focused on developing new deep learning architectures, this work proposes a systematic optimization framework for identifying practical low-channel EEG configurations by jointly analyzing channel selection, temporal segmentation, and preprocessing strategies for subject-independent motor imagery classification. This study investigates the influence of EEG channel reduction, temporal window segmentation, and baseline correction on motor imagery classification performance using EEGNet-based deep learning architectures. Experiments were conducted using the publicly available PhysioNet EEG Motor Movement/Imagery dataset under subject-independent evaluation conditions. Several EEG configurations were analyzed, including full-scale 64-channel recordings and reduced 15-, 6-, 3-, and 2-channel motor-cortex setups. The obtained results demonstrate that reduced-channel EEG configurations can achieve performance comparable to full-scale recordings. The best classification accuracy of 65.04% was achieved using a 15-channel motor configuration combined with 2 s sliding-window segmentation and baseline correction, achieving performance comparable to the conventional 64-channel setup (64.76%), while substantially reducing the number of electrodes and hardware complexity. Statistical analysis confirmed that the difference between the two configurations was not significant (paired t-test, p = 0.1684). Furthermore, compact 3-channel configurations maintained classification accuracy above 60%, supporting the feasibility of lightweight wearable EEG systems for practical BCI applications. The experiments additionally revealed that shorter temporal windows improve classification stability and reduce susceptibility to unrelated background EEG activity. Baseline correction significantly improved model generalization by compensating for inter-trial signal variability and slow EEG drift. Overall, the findings of this study demonstrate that careful optimization of electrode selection and preprocessing strategies can substantially improve the practicality of lightweight EEG-based motor imagery classification systems while reducing hardware complexity and preserving competitive performance. Full article

► Show Figures

Figure 1

17 pages, 3558 KB

Open AccessArticle

Vulnerability Detection Model Based on Clustering-Aware Heterogeneous Code Graphs

by Shize Lu, Lianmei Wang and Jiangtao Huang

Algorithms 2026, 19(7), 587; https://doi.org/10.3390/a19070587 - 16 Jul 2026

Viewed by 309

Abstract

Software vulnerabilities pose a severe threat to system security and stability. To reinforce the hierarchical structure of source code, enhance the distinctiveness of embedded features, and reduce feature confusion, we propose CADetect, a vulnerability detection model based on clustering-aware heterogeneous code graphs. First, [...] Read more.

Software vulnerabilities pose a severe threat to system security and stability. To reinforce the hierarchical structure of source code, enhance the distinctiveness of embedded features, and reduce feature confusion, we propose CADetect, a vulnerability detection model based on clustering-aware heterogeneous code graphs. First, a hierarchical clustering algorithm is applied to constrain and optimize the embedding vectors. This enhances the model’s ability to differentiate between diverse node and edge types within the heterogeneous code graphs, while streamlining code structures and reducing redundant information. Subsequently, multi-scale convolutions with varying kernel sizes are designed to extract vulnerability features at different granularities, effectively mitigating the information loss commonly caused by single-scale convolutions. Finally, extensive evaluations on the FFmpeg+Qemu, BigVul, and Reveal datasets demonstrate the effectiveness of the proposed model, and ablation studies confirm the specific contributions of the clustering algorithm and multi-scale convolutions. Full article

► Show Figures

Figure 1

31 pages, 12861 KB

Open AccessArticle

A Physics-Informed Explainable AI Framework for HVAC Anomaly Detection and Maintenance-Oriented Analysis in Urban Bus Fleets

by Bernardo Tormos, Ramón Sánchez-Márquez, Jorge Alvis-Sánchez and Vicente Bermudez

Algorithms 2026, 19(7), 586; https://doi.org/10.3390/a19070586 - 16 Jul 2026

Viewed by 336

Abstract

Although urban bus fleets increasingly generate HVAC operational data through onboard sensors and telematics platforms, their practical use for maintenance decision-making remains limited. In real-world conditions, these data are rarely accompanied by reliable fault labels, complete contextual information, or a clear physical interpretation [...] Read more.

Although urban bus fleets increasingly generate HVAC operational data through onboard sensors and telematics platforms, their practical use for maintenance decision-making remains limited. In real-world conditions, these data are rarely accompanied by reliable fault labels, complete contextual information, or a clear physical interpretation of the monitored variables. As a result, there is still a need for methodological frameworks capable of transforming imperfect operational data into physically meaningful and maintenance-oriented insights. This work proposes a structured framework that combines domain knowledge, unsupervised anomaly detection, and explainable artificial intelligence to support the initial exploitation of HVAC fleet data. The framework begins with sensor contextualization and physical mapping of available measurements onto the vapor-compression refrigeration cycle, enabling the construction of physically meaningful features and a coherent operational baseline. An Isolation Forest model is then used to identify deviations from this baseline, while Kernel SHAP explains the resulting deviation scores. SHAP contributions are further grouped into physics-informed explanation families, allowing anomalies to be interpreted in terms of HVAC mechanisms rather than isolated variables. The results show that this domain knowledge–XAI approach helps distinguish physically plausible rare operating states, inspection candidates, and ambiguous multivariate cases without relying on confirmed fault labels. The main contribution of this study is a physics-informed XAI framework that combines domain knowledge, unsupervised anomaly detection, and explainability methods to transform imperfect HVAC operational data from early-stage digitalized bus fleets into interpretable and maintenance-relevant information. Full article

(This article belongs to the Special Issue Explainable AI: Advances in Interpretability Algorithms and Applications)

► Show Figures

Figure 1

40 pages, 493 KB

Open AccessSystematic Review

The Evolution of Data Envelopment Analysis Models for Circular Economy Performance Assessment

by Andrey V. Lychev and Svetlana V. Ratner

Algorithms 2026, 19(7), 585; https://doi.org/10.3390/a19070585 - 16 Jul 2026

Viewed by 420

Abstract

Data Envelopment Analysis (DEA) has emerged as a major non-parametric technique for measuring efficiency in sustainability and environmental economics because it can handle multiple inputs and outputs without making explicit functional assumptions. DEA allows the simultaneous consideration of economic performance, resource utilization, environmental [...] Read more.

Data Envelopment Analysis (DEA) has emerged as a major non-parametric technique for measuring efficiency in sustainability and environmental economics because it can handle multiple inputs and outputs without making explicit functional assumptions. DEA allows the simultaneous consideration of economic performance, resource utilization, environmental impacts, and recycling results in the evaluation of the circular economy (CE). This review investigates the evolution of DEA models in the last years and the variables used to measure the CE performance. We analyze 209 peer-reviewed articles to systematically explore the evolution of DEA applications from conventional single-stage efficiency models to advanced network-based structures that better reflect the intricacy of circular systems. The review discusses the most advanced DEA approaches to date in the literature on CE assessment and uncovers specific factors that affect the choice of models in empirical studies. Finally, it points to promising directions of future research by showing interest in the development of comprehensive DEA models adapted to the specificity of the CE systems. Full article

(This article belongs to the Special Issue Data Envelopment Analysis for Decision Support)

► Show Figures

Figure 1

28 pages, 3422 KB

Open AccessArticle

Towards Explainable and Robust Cervical Cancer Screening Using Domain-Specific Transfer Learning Algorithm

by Jheelam Mondal, Mahendra Kumar Gourisaria, Rajdeep Chatterjee, Amitkumar V. Jha, Bhargav Appasani, Nicu Bizon and Cristian Toma

Algorithms 2026, 19(7), 584; https://doi.org/10.3390/a19070584 - 16 Jul 2026

Viewed by 362

Abstract

Cervical cancer is the fourth most frequent malignancy in women globally. Pap smear screening is important for early cancer detection, but manual smear analysis is time-consuming, labor-intensive and error-prone for diagnosis. Such issues in resource-limited areas have led to the introduction of deep [...] Read more.

Cervical cancer is the fourth most frequent malignancy in women globally. Pap smear screening is important for early cancer detection, but manual smear analysis is time-consuming, labor-intensive and error-prone for diagnosis. Such issues in resource-limited areas have led to the introduction of deep learning (DL) methods for automated cervical cancer diagnosis. But the majority of current methodologies depend on models pretrained on natural image datasets like ImageNet, which may inadequately represent domain-specific pathological characteristics. To mitigate this constraint, this research employs a domain-specific transfer learning algorithm approach using the PathMNIST histopathological dataset to enhance cervical cell classification. An accuracy score of 96.77% is achieved for the proposed YOLO* model, using the SIPAKMED dataset. To the best of available knowledge, no previous study has reported the use of PathMNIST as a pretraining source for cytology image classification. As domain-specific medical pretraining is becoming more popular, our study shows the importance of cross-domain generalization. Full article

(This article belongs to the Special Issue AI-Powered Biomedical Image Analysis)

► Show Figures

Figure 1

31 pages, 3755 KB

Open AccessArticle

The Evolution of Competitive Strategy: An Unsupervised Machine Learning Approach Using Topic Modeling and Keyword Clustering

by Cemal Zehir, Tuğçe Ekiz Yılmaz, Ali Kurt and Alex Borodin

Algorithms 2026, 19(7), 583; https://doi.org/10.3390/a19070583 - 16 Jul 2026

Viewed by 394

Abstract

The field of competitive strategy has attracted growing academic interest in recent years; however, the intellectual framework and thematic evolution of this research area remain fragmented. This study aims to systematically map the evolution of competitive strategy research using an unsupervised machine learning [...] Read more.

The field of competitive strategy has attracted growing academic interest in recent years; however, the intellectual framework and thematic evolution of this research area remain fragmented. This study aims to systematically map the evolution of competitive strategy research using an unsupervised machine learning framework. Drawing on a dataset of approximately 3900 journal articles indexed in the Scopus database between 2015 and 2025, the study employs probabilistic topic modeling, specifically Latent Dirichlet Allocation (LDA), together with keyword co-occurrence network analysis, thematic mapping, and community detection techniques to identify the latent thematic structure of the field. The findings reveal a modular and interconnected conceptual landscape in which capability-based strategic perspectives, particularly dynamic capabilities and the resource-based view, continue to occupy central positions in the literature. At the same time, themes related to digital transformation, artificial intelligence, supply chain resilience, environmental, social, and governance (ESG)-oriented management, and sustainability-focused strategic capabilities demonstrate substantial growth and emerging prominence. Temporal analyses further indicate a gradual reconfiguration toward digitally integrated, sustainability-oriented, and capability-driven strategic frameworks. By integrating topic modeling with network-based bibliometric analysis, the study provides a comprehensive and data-driven mapping of the field’s intellectual evolution. The study contributes to competitive strategy research by synthesizing the latent thematic structure of the field and by showing how complementary computational and bibliometric techniques can support large-scale literature mapping in a rapidly evolving research domain. Full article

► Show Figures

Figure 1

19 pages, 2675 KB

Open AccessArticle

Image Super-Resolution Reconstruction Based on Hierarchical Feature Aggregation and Laplacian High-Frequency Compensation

by Kangliang Xiao, Shaozhang Xiao, Bolun Chen, Yuanyuan Wang and Raees ul Haq Muhammad

Algorithms 2026, 19(7), 582; https://doi.org/10.3390/a19070582 - 16 Jul 2026

Viewed by 258

Abstract

Existing image super-resolution methods still suffer from limitations in edge-structure restoration, high-frequency texture preservation, and artifact suppression, which may lead to blurred contours and unnatural textures. To address these issues, this paper proposes an image super-resolution method based on hierarchical feature aggregation and [...] Read more.

Existing image super-resolution methods still suffer from limitations in edge-structure restoration, high-frequency texture preservation, and artifact suppression, which may lead to blurred contours and unnatural textures. To address these issues, this paper proposes an image super-resolution method based on hierarchical feature aggregation and Laplacian high-frequency compensation. First, a Hierarchical Feature Aggregation Attention Block (HFAB) is designed in the generator to progressively extract image features at different levels through multiple convolutional layers. A High-Frequency Variance Adaptive Channel Attention Block (HFVB) is further introduced to adaptively enhance key texture and edge information. Second, a Laplacian Adaptive Upsampling (LAU) module is developed to combine low-frequency content reconstruction with high-frequency detail compensation, thereby strengthening edge contours, preserving fine textures, and reducing artifacts. Finally, a Dissimilarity Structural Similarity Index Measure (DSSIM) loss is incorporated into the loss function to constrain local structural consistency and further improve the structural preservation and perceptual quality of reconstructed images. Experimental results on Set5, Set14, BSD100, and Urban100 show that, compared with SRGAN, the proposed method improves PSNR by 0.31 dB, 0.18 dB, 0.14 dB, and 0.14 dB, respectively, while reducing LPIPS by 0.0234, 0.0183, 0.0222, and 0.0222. These results indicate that the proposed method provides consistent improvements over SRGAN and achieves modest, metric-dependent gains over ESRGAN, suggesting an incremental enhancement in reconstruction accuracy and perceptual quality on both natural image benchmarks and complex urban scene datasets. Full article

(This article belongs to the Special Issue Machine Learning Algorithms for Image Understanding and Analysis (2nd Edition))

► Show Figures

Figure 1

22 pages, 447 KB

Open AccessArticle

Kalman-Annealing: Calibrated Uncertainty for Simulated Annealing via a Probabilistic-Numerics Filter, with an Application to Reinforcement-Learning Hyperparameter Tuning

by Eduardo C. Garrido-Merchán

Algorithms 2026, 19(7), 581; https://doi.org/10.3390/a19070581 - 15 Jul 2026

Viewed by 270

Abstract

Noisy, expensive, gradient-free optimisers—simulated annealing chief among them—almost never report how confident one should be in the configuration they return, and reinforcement-learning hyperparameter tuning, where the noise is large and the budget tight, is the setting where this silence hurts most. The contribution [...] Read more.

Noisy, expensive, gradient-free optimisers—simulated annealing chief among them—almost never report how confident one should be in the configuration they return, and reinforcement-learning hyperparameter tuning, where the noise is large and the budget tight, is the setting where this silence hurts most. The contribution of this paper is a mechanism for uncertainty quantification, not a faster optimiser: we equip simulated annealing with a calibrated credible interval over the value of the recovered configuration, and we are explicit that this comes at an optimisation cost that only some landscapes repay. We introduce Kalman-Annealing (KA), a minimal modification of simulated annealing in which a one-dimensional Kalman filter—the canonical probabilistic numerical method—is interleaved with the Metropolis acceptance step. The filter denoises each return before acceptance, and a short terminal refinement of the best visited state converts the run into a calibrated credible interval over the value of the recovered hyperparameter. A single analytical identity,

Q_{t} = c T_{t}^{2}

, couples the filter process noise to the cooling schedule and absorbs the only free parameter of the filter into one already present in the metaheuristic. Under standard cooling assumptions the credible intervals are calibrated and the posterior variance contracts at a rate compatible with simulated-annealing convergence. On synthetic benchmarks (a noisy five-dimensional quadratic and the noisy Branin function, 200 seeds each) and on hyperparameter tuning of REINFORCE on three classic-control tasks (10 seeds each), the empirical

90 %

coverage of KA’s credible intervals lies within sampling error of the nominal level—a property none of the baselines provides—and the optimiser overhead is close to four orders of magnitude below that of Gaussian-process Bayesian optimisation. The interval cannot be extracted for free from an unmodified SA run: an interval built from the trailing evaluations of the vanilla trajectory fails to calibrate in every reading we test, and the repair that does calibrate is exactly KA’s terminal-refinement phase grafted onto the unfiltered chain, at the same cost in diverted evaluations. Honest scoreboard: On simple regret, KA is at best on par with vanilla simulated annealing on the unimodal synthetic (the nominal advantage does not survive correction for multiple comparisons) and loses to the SA family, to CMA-ES and to Gaussian-process Bayesian optimisation on the multi-modal and ill-conditioned synthetics and on the informative reinforcement-learning tasks, under both REINFORCE and PPO. We trace this gap quantitatively to the filter acting as a low-pass, with a mean Kalman gain near one half, on the favourable-tail observations that drive SA’s basin escape, and we delineate the operating regime in which the calibrated-uncertainty contribution of KA is worth its optimisation cost. Full article

(This article belongs to the Special Issue Metaheuristic Algorithms in Optimal Design of Engineering Problems (2nd Edition))

► Show Figures

Figure 1

23 pages, 3127 KB

Open AccessArticle

A Two-Stage Diagnostic Framework for Winner-Takes-All Failures Reveals Hollow Diversity in Long-Tail Trajectory Prediction

by Zimo Zhuang

Algorithms 2026, 19(7), 580; https://doi.org/10.3390/a19070580 - 15 Jul 2026

Viewed by 273

Abstract

The Winner-Takes-All (WTA) training paradigm inherently prioritises global trajectory accuracy. For long-tail samples, where terminal prediction is inherently difficult, this prioritisation leads to degraded Final Displacement Error (FDE). This asymmetry is reflected in the considerably poorer performance of FDE than Average Displacement Error [...] Read more.

The Winner-Takes-All (WTA) training paradigm inherently prioritises global trajectory accuracy. For long-tail samples, where terminal prediction is inherently difficult, this prioritisation leads to degraded Final Displacement Error (FDE). This asymmetry is reflected in the considerably poorer performance of FDE than Average Displacement Error (ADE) on such samples, yet its underlying mechanisms remain insufficiently diagnosed in the literature. The core contribution of this paper is a two-stage diagnostic framework that reveals two deeper asymmetries and identifies a previously unrecognised phenomenon, which we term hollow diversity. The framework is designed to be independent of specific model architectures and datasets. We first deployed the framework on the JAAD dataset using a custom-implemented baseline and subsequently validated its effectiveness on the ETH/UCY datasets. Full article

(This article belongs to the Section Evolutionary Algorithms and Machine Learning)

► Show Figures

Figure 1

25 pages, 10533 KB

Open AccessArticle

IEC 60601-2-40-Based Evaluation of an Embedded sEMG Platform for Muscle Activation Analysis

by Lester Fitz Gama, Juvenal Rodríguez-Reséndiz, Francisco David Pérez Reynoso, César Omar Capetillo Contreras and Luis Alberto Gordillo Roblero

Algorithms 2026, 19(7), 579; https://doi.org/10.3390/a19070579 - 15 Jul 2026

Viewed by 236

Abstract

Surface electromyography (sEMG) is widely used to evaluate neuromuscular activity; however, objective and reproducible methodologies for assessing signal quality in embedded acquisition systems remain limited. This study presents a multichannel embedded sEMG acquisition platform based on an ADS1298 analog front-end (Texas Instruments, Dallas, [...] Read more.

Surface electromyography (sEMG) is widely used to evaluate neuromuscular activity; however, objective and reproducible methodologies for assessing signal quality in embedded acquisition systems remain limited. This study presents a multichannel embedded sEMG acquisition platform based on an ADS1298 analog front-end (Texas Instruments, Dallas, TX, USA) and an STM32H743ZIT6 microcontroller (STMicroelectronics, Geneva, Switzerland), together with a signal-quality evaluation methodology guided by signal-integrity principles derived from IEC 60601-2-40. sEMG signals were acquired at 2 kHz from 10 healthy participants during standardized submaximal isometric and controlled isotonic contractions using consistent electrode placement and acquisition procedures. Signal quality was quantified using complementary temporal and spectral metrics, including signal-to-noise ratio (SNR), power-line interference ratio (PLI), baseline drift, root mean square stability, median frequency, spectral entropy, skewness, and kurtosis. Signal conditioning reduced the baseline drift from 0.32 ± 0.23 to 0.004 ± 0.003 and reduced PLI from 0.097 ± 0.135 to 0.010 ± 0.003 while preserving contraction-dependent temporal and spectral characteristics. In addition, envelope-based Spearman correlation analysis revealed contraction-dependent intermuscular coordination patterns between the long and short heads of the biceps brachii under controlled acquisition conditions. These findings demonstrate that combining standardized acquisition protocols, objective signal quality metrics, and interpretable correlation-based analysis provides a reproducible engineering framework for evaluating embedded sEMG systems and establishes a foundation for future machine learning approaches based on larger and well-characterized physiological datasets. Full article

(This article belongs to the Special Issue Machine Learning Techniques for Brain Data Analysis Using EEG, EMG or Image Data)

► Show Figures

Figure 1

25 pages, 4257 KB

Open AccessArticle

A Numerov–Galerkin Framework for the Transient Dynamics of Anisotropic Plates on Vlasov Foundations

by Adebola Samuel Adeoye, Ezekiel Olaoluwa Omole, Babatope Omolofe, Taiwo Stephen Fayose and Aseel Smerat

Algorithms 2026, 19(7), 578; https://doi.org/10.3390/a19070578 - 15 Jul 2026

Viewed by 269

Abstract

In this study, a high-order Galerkin–Numerov approach is presented to solve the transient vibration problem of anisotropic Kirchhoff plates supported by a uniform Vlasov foundation. A discretization of the governing fourth-order plate equation is derived based on a mixed boundary value problem and [...] Read more.

In this study, a high-order Galerkin–Numerov approach is presented to solve the transient vibration problem of anisotropic Kirchhoff plates supported by a uniform Vlasov foundation. A discretization of the governing fourth-order plate equation is derived based on a mixed boundary value problem and a hybrid Hermite–sine Galerkin formulation, which maintains the C1-continuity properties of classical plate theory. The resulting reduced-order modal system is integrated in time with the Numerov scheme, which is fourth-order accurate, and has a small numerical dispersion and good phase-preserving properties for oscillatory dynamics. The proposed methodology is evaluated using stability and convergence tests and parametric investigations. The fourth-order temporal convergence and rapid spectral-like spatial convergence of the numerical results are validated, and the long-time accuracy and robustness of the formulation is confirmed by the negligible phase error and bounded energy drift. The results from the parametric study indicate that the thickness of the plates and the stiffness of the Winkler foundation are the two most important mechanisms for vibration suppression, while the orthotropic coupling and the Vlasov shear interaction have substantial effects on the modal redistribution and transient deformation properties. The proposed method is compared with the conventional lower-order integration schemes, and it is observed that the method gives better phase fidelity and computational efficiency, and it is possible to predict the vibration amplitude and vibration timing accurately. In addition to the numerical benefits, the framework also provided physical insights on the coupled effect of anisotropy, foundation interaction and boundary restraint. The suggested model is directly applicable for composite floor systems, aerospace panels, foundation supported slabs, biomechanical plate analogs, etc., and smart vibration control platforms. This work thus lays the groundwork for future studies of nonlinear behavior, adaptive foundations and digital twin simulation of structural systems and presents a strong and scalable computational tool for the study of plate–foundation dynamics. Full article

(This article belongs to the Section Algorithms for Multidisciplinary Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Algorithms, Volume 19, Issue 7 (July 2026) – 106 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI