Abstract
The integration of Deep Learning techniques for grapevine yield estimation has led to significant advancements in Precision Viticulture. The accurate detection and counting of berries per bunch is a critical task that can explain up to 30% of yield variability, thereby enabling improved yield estimation. This study proposes a YOLO-based approach for the automated detection and counting of visible grapevine berries, using a dataset of more than 1500 images collected over three phenological stages. The selected YOLO models performed well in both detection and counting tasks, with all models achieving high detection accuracy (G-mAP ' 0.95) and estimation of visible berries (R2 ' 0.97). Among the evaluated models, YOLOv11n exhibited the highest detection performance (F1-Score = 0.954, G-mAP = 0.962), while YOLOv10n demonstrated the most consistent and reliable counting accuracy (MAPE = 4.764, MSE = 12.203, RMSE = 3.493). Beyond overall performance, the analysis revealed that ampelographic features such as berry size, occlusion, and bunch morphology can influence accuracy, although YOLOv10n showed no significant disparities across categories. To extend the scope, a complementary analysis demonstrated a strong linear relationship (R2 = 0.860) between visible counts and the total number of berries per bunch, supporting the potential of correction models to address occlusion. By systematically evaluating model behaviour across diverse viticultural conditions and incorporating correlation with total berry counts, this study provides a deeper understanding of the robustness and limitations of Deep Learning models, offering critical insights for future applications in vineyard monitoring, yield estimation, harvest optimisation, and management. © 2026 The Authors.

Abstract
In the evolving field of machine learning, ensuring group fairness has become a critical concern, prompting the development of algorithms designed to mitigate bias in decision-making processes. Group fairness refers to the principle that a model's decisions should be equitable across different groups defined by sensitive attributes such as gender or race, ensuring that individuals from privileged groups and unprivileged groups are treated fairly and receive similar outcomes. However, achieving fairness in the presence of group-specific concept drift remains an unexplored frontier, and our research represents pioneering efforts in this regard. Group-specific concept drift refers to situations where one group experiences concept drift over time, while another does not, leading to a decrease in fairness even if accuracy (ACC) remains fairly stable. Within the framework of federated learning (FL), where clients collaboratively train models, its distributed nature further amplifies these challenges since each client can experience group-specific concept drift independently while still sharing the same underlying concept, creating a complex and dynamic environment for maintaining fairness. The most significant contribution of our research is the formalization and introduction of the problem of group-specific concept drift and its distributed counterpart, shedding light on its critical importance in the field of fairness. In addition, leveraging insights from prior research, we adapt an existing distributed concept drift adaptation algorithm to tackle group-specific distributed concept drift, which uses a multimodel approach, a local group-specific drift detection mechanism, and continuous clustering of models over time. The findings from our experiments highlight the importance of addressing group-specific concept drift and its distributed counterpart to advance fairness in machine learning.

Abstract
This work addresses the joint challenges of layout design and power output smoothing in Airborne Wind Energy farms. The problem is deciding the power units' location and their synchronization, as well as balancing the power output in Airborne Wind Energy Farms, while maximizing the average power output. Airborne Wind Energy systems, particularly the ground-generation type, are susceptible to significant power output fluctuations, which must be mitigated to ensure a stable power flow injected into the electrical grid. We propose and evaluate various synchronization strategies, along with an innovative retraction method introduced in this work, which eliminates the need for synchronized operational cycles among kites. This approach increases flexibility in production cycle selection, leading to a significantly smoother power output signal.

Abstract

Abstract
Large Events Models (LEMs) are a class of models designed to predict and analyze the sequence of events in soccer matches, capturing the complex dynamics of the game. The original LEM framework, based on a chain of classifiers, faced challenges such as synchronization, scalability issues, and limited context utilization. This paper proposes a unified and scalable approach to model soccer events using a tabular autoregressive model. Our models demonstrate significant improvements over the original LEM, achieving higher accuracy in event prediction and better simulation quality, while also offering greater flexibility and scalability. The unified LEM framework enables a wide range of applications in soccer analytics that we display in this paper, including real-time match outcome prediction, player performance analysis, and game simulation, serving as a general solution for many problems in the field.

Abstract
Abstract While automation has transformed many areas inside clinical laboratories, microbiology still relies heavily on manual tasks, particularly the culture of samples on agar plates and their subsequent manual review for microorganism identification and antibiotic susceptibility profiling. Bacterial colony detection and classification require trained professionals, making the process time-consuming and prone to human error. Developing deep learning models to automate these tasks could improve microbiology workflows and accelerate clinical decision-making. In this study we trained and evaluated five object detection architectures (Faster R-CNN and RetinaNet with ResNet-50 and ResNet-101 backbones, and YOLOv8) on the Annotated Germs for Automated Recognition (AGAR) dataset for bacterial colony classification. Transfer learning, cross-subset generalization, and Weighted Box Fusion (WBF) ensemble methods were applied to enhance and characterize performance. Additionally, we created and publicly released a curated dataset of 165 agar plate images containing colonies of S. aureus , P. aeruginosa , and E. coli cultured across four distinct culture media. YOLOv8m achieved a mean Average Precision (mAP) of 69.0% on the AGAR dataset, outperforming the best Detectron2 model (Faster R-CNN ResNet-101, 63.1%) by 5.9 percentage points. A four-model WBF ensemble combining both architectures reached 70.5% mAP (95% CI: 68.4–71.7). Cross-subset evaluation showed that a single model trained on the full dataset generalizes well to individual imaging conditions, making subset-specific fine-tuning largely unnecessary. On the curated dataset, a mixed ensemble reached 58.7% mAP (95% CI: 57.1–63.7). These results demonstrate that architecture choice and training data diversity are the primary drivers of performance for colony detection on agar plates.