Abstract
The rising adoption rates and integration of Renewable Energy Sources (RES) and Electric Vehicles (EVs) into the energy grid introduces complex challenges, including the need to balance supply and demand and smooth peak consumptions. Addressing these challenges requires innovative solutions such as Demand Response (DR), Renewable Energy Communities (RECs), and more specifically for EVs, Vehicle-to-Grid (V2G). However, existing V2G approaches often fall short in real-world applicability, adaptability, and user engagement. To bridge this gap, this paper proposes EnergAIze, a Multi-Agent Reinforcement Learning (MARL) energy management algorithm leveraging the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm. EnergAIze enables user-centric multi-objective energy management by allowing each prosumer to select from a range of personal management objectives. Additionally, it architects' data protection and ownership through decentralized deployment, where each prosumer can situate an energy management node directly at their own dwelling. The local node not only manages local EVs and other energy assets but also fosters REC wide optimization. EnergAIze is evaluated through a case study using the CityLearn framework. The results show reduction in peak loads, ramping, carbon emissions, and electricity costs at the REC level while optimizing for individual prosumers objectives.

Abstract
Keratitis is an inflammatory corneal condition responsible for 10% of visual impairment in low- and middleincome countries (LMICs), with bacteria, fungi, or amoeba as the most common infection etiologies. While an accurate and timely diagnosis is crucial for the selected treatment and the patients' sight outcomes, due to the high cost and limited availability of laboratory diagnostics in LMICs, diagnosis is often made by clinical observation alone, despite its lower accuracy. In this study, we investigate and compare different deep learning approaches to diagnose the source of infection: 1) three separate binary models for infection type predictions; 2) a multitask model with a shared backbone and three parallel classification layers (Multitask V1); and, 3) a multitask model with a shared backbone and a multi-head classification layer (Multitask V2). We used a private Brazilian cornea dataset to conduct the empirical evaluation. We achieved the best results with Multitask V2, with an area under the receiver operating characteristic curve (AUROC) confidence intervals of 0.7413-0.7740 (bacteria), 0.83950.8725 (fungi), and 0.9448-0.9616 (amoeba). A statistical analysis of the impact of patient features on models' performance revealed that sex significantly affects amoeba infection prediction, and age seems to affect fungi and bacteria predictions.

Abstract
Effective navigation in mobile robotics relies on precise environmental mapping, including the detection of complex objects as geometric primitives. This work introduces a deep learning model that determines the pose, type, and dimensions of 2D primitives using a mobile robot equipped with a noisy LiDAR sensor. Simulated experiments conducted in Webots involved randomly placed primitives, with the robot capturing point clouds which were used to progressively build a map of the environment. Two mapping techniques were considered, a deterministic and probabilistic (Bayesian) mapping, and different levels of noise for the LiDAR were compared. The maps were used as input to a YOLOv5 network that detected the position and type of the primitives. A cropped image of each primitive was then fed to a Convolutional Neural Network (CNN) that determined the dimensions and orientation of a given primitive. Results show that the primitive classification achieved an accuracy of 95% in low noise, dropping to 85% under higher noise conditions, while the prediction of the shapes' dimensions had error rates from 5% to 12%, as the noise increased. The probabilistic mapping approach improved accuracy by 10-15% compared to deterministic methods, showcasing robustness to noise levels up to 0.1. Therefore, these findings highlight the effectiveness of probabilistic mapping in enhancing detection accuracy for mobile robot perception in noisy environments.

Abstract
This paper focuses on the importance of Business Intelligence (BI) tools in the business context and the urgent need for more effective implementation of time series forecasting models in these resources. It shows the utility and applicability of Sage Enterprise Intelligence (SEI), an integrated BI tool in Enterprise Resource Planning (ERP) Sage, by illustrating how it enhances data analysis and decision-making processes. Additionally, a study will show the application of time series forecasting models: Seasonal AutoRegressive Integrated Moving Average (SARIMA) and additive Holt-Winters to the sales value of a fuel sector company. The research was conducted through a case study in which sales data were collected from 2016 to 2023. The results indicate that neither of the two models exceeded the sales figures reflecting the company's market position. In this case study, both models performed well, with the residuals verifying the assumptions. However, the additive Holt-Winters model had lower errors, which is why it was selected for the final step: forecasting 12 months.

Abstract
Recently, multi-carrier energy systems (MCESs) have been rapidly developed as flexible multi-generation systems aiming to satisfy load demands by purchasing, converting, and storing different energy carriers. This study specifically focuses on the optimal and robust large-scale coordination of interconnected energy hubs (IEHs) in an iterative consensus-based procedure considering distribution network losses. Furthermore, a new robustbased hybrid IGDT/consensus algorithm is introduced to achieve risk-averse optimal energy management in IEHs under uncertainty. The fast convergence, needless to collect the total information from all hubs, minimal computational burden, and more robust communication system are the most important features of the proposed distributed consensus algorithm in this study. The effectiveness of the proposed consensus algorithm is verified by simulation results considering various energy trading structures in IEHs at different scales. The obtained results highlight the scalability capability of the proposed method. Regarding an IEHS of 30 energy hubs, the computation burden is lightened by 0.53 (s) and 0.1917 (s), respectively with and without uncertainty. Considering distribution network losses, the total purchasing costs can be increased by 8%. The simulation results also reveal an increase of 11% in the total power trading under the uncertainty.

Abstract
The scoping review reported by this paper aimed to analyze and synthesize state-of-the-art studies focused on the application of machine learning methods to enhance the cyber resilience of cyber-physical systems. An electronic search was conducted, and 24 studies were included in this review after the selection process. The most representative application domains were computer networks and power systems, while in terms of cyber resilience functions, risk identification, risk mitigation or protection, and detection of anomalous situations were the most implemented functions. Moreover, the results of this scoping review show that the interest in the topic of cyber resilience and machine learning is quite recent, which justifies the heterogeneity of the included studies in terms of machine learning methods and datasets being used for the experimental validations, as well as in terms of outcomes being measured. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.