Handwritten digit recognition (HDR) remains challenging due to variations in writing styles. To address this challenge, this study comprehensively compares ML (ML) and deep learning (DL) models. We explored a variety of approaches. We evaluated these models on the Modified National Institute of Standards and Technology (MNIST) and Extended Modified National Institute of Standards and Technology (EMNIST) datasets to assess their generalization capabilities. Initially, we investigated standalone ML and DL models trained from scratch to learn features directly. Logistic regression (LR) achieved an accuracy of 92.5% on MNIST and 86.63% on EMNIST. A multi-layer perceptron demonstrated improved performance with 98.10% accuracy on MNIST. Convolutional neural networks exhibited superior performance, reaching 99.90% accuracy on MNIST and 99.57% on EMNIST. To further enhance performance, we explored ensemble learning techniques, combining CNNs with RF (98.20 and 99.86% accuracy on MNIST and EMNIST, respectively), LR (88.67 and 99.79% accuracy on MNIST and EMNIST, respectively), and VC (99.27 and 99.83% accuracy on MNIST and EMNIST, respectively). We then introduced a ViT model, leveraging self-attention for long-range dependency modeling, achieving an accuracy of 98.70% on MNIST and 99.58% on EMNIST. Finally, we proposed a hybrid model combining CNN and ViT, that yielded the highest accuracy of 99.97% on MNIST and 98.26% on EMNIST. Throughout our experimentation, we employed various techniques such as regularization, weight initialization, and optimization strategies to improve model performance. The impact of each technique is analyzed and discussed. Overall, this study provides a comprehensive comparison of different HDR models, highlighting each approach’s strengths and weaknesses. The results demonstrate the effectiveness of DL models, particularly CNNs and hybrid architectures, in achieving high accuracy in HDR.

Demand-side management (DSM) is a security-critical function in residential smart grids. The same communication and sensing infrastructure that enables fine-grained load flexibility also exposes schedulers to corrupted measurements, price manipulation, and delayed control signals. Conventional DSM formulations generally treat cyber and communication impairments as external disturbances, which are addressed only after the schedule has already been calculated. This study proposes and evaluates Cyber-Resilient and QoS-Aware Demand-Side Management (CQ-DSM) as a hierarchical optimization framework that embeds cyber-risk likelihood and communication quality-of-service (QoS) directly into the scheduling objective. Local home energy management systems (HEMSs) solve mixed-integer linear programs at the appliance level, and central aggregators broadcast compact coordination signals based on real-time prices, measured QoS, and a sliding-window GRU-feature MLP risk estimator. The key intuition is to convert uncertainty about trust and actuation reliability into scheduling prices: high cyber risk discourages exposed loads during vulnerable periods, whereas poor QoS increases the value of locally preserving thermal flexibility. Under the simulation conditions (NYISO August pricing, P = 50 prosumers, Seed 42), CQ-DSM reduces overall system costs by 5.75% and imbalance procurement costs relative to an attack-unaware baseline under normal operation, limits the FDI-induced cost increase to 0.46% versus 0.83% (44% reduction in cost overrun), and reduces thermal-violation penalties by 81% under degraded QoS. The ablation results are consistent with cyber-risk pricing and QoS-aware fallback being complementary rather than redundant under the scenarios tested.

Logo classification is crucial in various applications, including brand monitoring, copyright protection, and digital forensics. Traditional computer vision techniques face significant limitations, particularly in handling scale variations, occlusions, and background clutter. While deep learning models, particularly convolutional neural networks (CNNs), offer superior solutions, they often come with high computational costs, posing challenges for real-time deployment. This paper introduces LogoXpertNet, a lightweight deep learning architecture specifically designed for efficient logo classification. The key innovations of LogoXpertNet include: (1) a modified MobileNetV3 backbone enhanced with bottleneck and squeeze-and-excitation (SE) blocks for efficient feature extraction; (2) a novel cross-layer feature fusion (CLFF) module that improves feature integration across network depths; (3) a newly proposed hierarchical squeeze-excitation spatial attention block (HSE-SAB) that dynamically attends to both spatial and channel-wise features; and (4) a feature-aware convolution block attention module (FA-CBAM) that uniquely fuses spatial and frequency-domain information for refined logo attention. Extensive experiments on the evaluated benchmark datasets, including FlickrLogos-32, BelgaLogos, and WebLogo-2M, show that LogoXpertNet achieves strong classification performance while maintaining low computational overhead. Because the reported accuracies approach saturation on these benchmarks, the results should be interpreted in the context of dataset characteristics, split construction, and metric definition rather than as evidence of universal performance across all real-world logo-recognition scenarios. LogoXpertNet provides an efficient and practical framework for benchmark logo classification and offers a promising basis for future validation under more challenging real-world conditions.

Internet of Things (IoT) is composed of many IoT devices connected throughout the Internet, that collect and share information to represent the environment. IoT is currently restructuring the actual manufacturing to smart manufacturing. However, inherent characteristics of IoT lead to a number of titanic challenges such as decentralization, weak interoperability, security, etc. The artificial intelligence provides opportunities to address IoT’s challenges, e.g the agent technology. This paper presents first an overview of ML and discusses some related work. Then, we briefly present the classic IoT architecture. Then we introduce our proposed Intelligent IoT (IIoT) architecture. We next concentrate on introducing the approach using multi-agent DRL in IIoT. Finally, in this promising field, we outline the open directions of future work.

Urban services increasingly depend on interconnected sensing, communication, computing, decision-support, and response functions, yet technical performance alone does not establish service effectiveness. This structured integrative review examines WSNs, WBANs, V2X systems, 5G-enabled edge–cloud infrastructures, and prospective 6G capabilities across health, mobility, environmental monitoring, energy, water, infrastructure safety, and climate resilience. A five-database search covering January 2018–March 2025 was supplemented by citation tracing and a documented gap-directed update with a final cutoff of 1 July 2026. The analytical corpus comprised 40 peer-reviewed studies, five deployment cases, and one scope-boundary case. An outcome-mediated perception–network–edge/cloud–decision–response framework enabled categorical comparison of heterogeneous evidence without pooling non-comparable measures. Attribution was classified as T1 (comparatively evaluated downstream outcome), T2 (technical or bounded operational outcome), or T3 (conceptual linkage): three studies were T1, 34 T2, and three T3. Ten studies supported target-level SDG alignment, whereas none reached indicator-level correspondence. Evidence remained concentrated at communication, processing, decision-support, and bounded operational endpoints, with limited assessment of response availability and disruption–recovery conditions. Among the deployment cases, only SFpark supported T1 interpretation. These findings characterize the selected corpus and identify a persistent gap between technical performance and comparative, longitudinal, and distributionally assessed urban-service outcomes.

Antimicrobial resistance (AMR) poses a critical global challenge, driven by the widespread misuse of antibiotics across human, animal, and environmental domains. Conventional laboratory-based antimicrobial susceptibility testing is time-consuming and resource-intensive, limiting its scalability and clinical responsiveness. Artificial intelligence (AI) and deep learning (DL) have emerged as effective computational tools for predicting AMR from heterogeneous data sources. This scoping review systematically maps AI- and DL-based methodologies for AMR prediction across 20 peer-reviewed studies published between 2021 and 2025. Following PRISMA-ScR guidelines, we analyze methodological trends, strengths, and limitations across three thematic clusters: genomic and spectrometric prediction, clinical and EHR-based prediction, and proteomic antibiotic resistance gene detection. A three-generation architectural progression is observed, from classical machine learning and CNN-based models, through XAI-integrated CNNs functioning as biological discovery tools, to large language model and foundation model approaches enabling few-shot adaptation to emerging antibiotics. Seven distinct explainability methods are identified, of which four achieve formally validated attributions and five demonstrate biologically grounded explanations, with partial overlap between categories. Key challenges include limited data standardization, absence of temporal and geographic external validation, heterogeneous and unvalidated explainability practices, and real-world deployment constraints. This review provides a structured technical synthesis of AI-driven AMR prediction methodologies and outlines future research priorities toward multi-species evaluation, temporally validated models, standardized explainability reporting, and clinically deployable systems aligned with the One Health paradigm.