Search Results (1,703)

Federated learning (FL) has emerged as an important distributed machine learning paradigm allowing many users to train a model together without sharing raw data. However, FL’s decentralized design makes it very susceptible to both Byzantine and coordinated model poisoning attacks, where a few malicious rounds of gradients are strategically inserted to reduce the overall integrity of the model. Even the existing Byzantine-resilient aggregation methods such as geometric median, Krum, and trimmed mean are vulnerable to sophisticated, coordinated poisoning attacks that leverage statistical gaps in outlier detection. In this article, we present FedSentinel, a novel Byzantine-resilient federated learning framework that combines cryptographic gradient attestation with adaptive trust-weighted aggregation to protect against coordinated model-poisoning attacks, which are among the most serious challenges. Three key innovations are introduced in FedSentinel Cryptographic Gradient Attestation Protocol (CGAP) that utilizes commitment schemes and zero-knowledge range proofs to ensure the integrity of the gradients and verify that submitted updates satisfy predefined norm and direction constraints; in the current protocol configuration, CGAP provides gradient integrity verification rather than full gradient confidentiality from the server Dynamic Trust-Weighted Robust Aggregation (DT-RoA): the computation and updating of per-client trust scores based on historical gradient consistency and cross-validation signals; and Coordinated Attack Detection Engine (CADE): based on spectral analysis of the gradient covariance matrices, it detects and isolates colluding Byzantine clients. By participating in up to 100 clients on CIFAR-10, CIFAR-100, FEMNIST, and Sentiment140 datasets, extensive experiments conducted under up to 30% Byzantine adversaries reveal that FedSentinel achieves 91.36% average global accuracy, whereas state-of-the-art defenses such as FLTrust, FLAME, RoFL, ShieldFL, and DnC achieve 83.83–86.94%. FedSentinel outperforms these defenses by 4.42–7.53% in terms of accuracy, while decreasing attack success rates by 53.2% under coordinated Byzantine backdoor attacks. The proposed framework offers a promising approach for federated learning that is verifiable and trustworthy in adversarial environments. Full article

Internet of Vehicles (IoV) intrusion detection systems (IDSs) require collaborative learning that preserves raw-data locality while producing independently checkable post-run evidence. This paper presents FL-BC-IDS, an evidence-native, privacy-aware hierarchical federated IDS in which vehicles train Differentially Private XGBoost models, roadside units perform deterministic admission and tree-bagging aggregation, and the GLOBAL stage forms an equal-weight ensemble over validated RSU models. Signed reports, privacy records, SHA-256/Poseidon commitments, scoped Groth16 proofs, reconstructable public inputs, and digest-pinned blockchain receipts provide a unified verification path. Across 10 seed-controlled runs, the mean±SD accuracy/F1 values were $0.998021\pm 0.000246$/$0.983597\pm 0.002053$ on CSE-CIC-IDS2018 and $0.999867\pm 0.000152$/$0.999495\pm 0.000579$ on CICIoV2024. With thresholds fixed exclusively from development data, the strict held-out-attack macro recall was 0.8031 and 0.9090 on CSE-CIC-IDS2018 and CICIoV2024, respectively, indicating residual attack-specific generalization limitations; supervised rolling-origin temporal refresh on CSE-CIC-IDS2018 achieved 0.984788 pooled seen-attack recall at a 0.005700 test FPR. A controlled 20-vehicle, eight-round heterogeneity and participation stress test retained 0.998151 accuracy and 0.984782 F1-score. Verification rejected invalid or context-mismatched artifacts and independently checked model–anchor consistency, RSU aggregation replay, commitments, and public inputs. The reported DP budgets are conditional learner-stage bounds for learner-input record instances, not end-to-end guarantees for original pre-preprocessing records. Full article

While federated learning offers a decentralized approach to model training, ensuring the integrity of the information from each IoT client remains a challenge. This work delves into the dynamics of multi-stage federated learning, its susceptibility to information integrity attacks, and how to defend against such threats. A comprehensive understanding of data uncertainty and the challenges of poisoning attacks is discussed, laying a solid groundwork for the proposed defense mechanisms. At its core, this paper introduces a novel multi-stage federated learning model that segments the federated learning process into distinct phases with a novel approach of inertial momentum-aware Inexact Augmented Lagrange Multiplier Robust PCA with constant momentum factor and unaltered norm of the traditional one, each tailored to optimize for both efficiency and security. This robust framework is then tested against data injection-based poisoning attacks, using sparse noise, and demonstrates the effectiveness of the proposed recovery techniques like Robust PCA. Performance results highlight the resilience and efficiency of the introduced model with novel reconstruction algorithm, emphasizing the importance of this approach in real-world IoT settings. Data analysis, model summaries, and impacts of adversarial attacks further reinforce the findings, which are evaluated using rigorous statistical metrics and machine learning algorithms. The paper concludes by acknowledging its efficiency in detection and recovery from data poisoning attacks, improving robustness and data reconstruction in IoT environments while highlighting opportunities for further security enhancements. Full article

The integration of digital therapeutics (DTx), wearable electronic devices, and artificial intelligence (AI) represents a significant advancement in personalized healthcare. The primary purpose of this structured narrative review is to evaluate the convergence of these technologies, providing a consolidated framework that bridges the gap between raw biometric data acquisition and actionable, AI-driven clinical insights. This paper synthesizes the latest literature on the intersection of mobile health (mHealth), machine learning (ML), and physiological tracking, with a primary focus on heart rate variability (HRV) and associated biochemical markers, such as cortisol, salivary alpha-amylase, and interleukins. Instead of viewing wearable outputs simply as raw data, we critically evaluate the technical verification and clinical validation required to define them as true “digital biomarkers.” By evaluating multimodal sensor technologies and advanced predictive algorithms, this paper outlines the clinical utility of digital biomarkers in diagnosing and proactively managing cardiovascular, neurological, metabolic, and psychiatric conditions, noting classification accuracies frequently exceeding 85% in controlled settings. However, we strongly caution that internally validated performance in controlled settings does not inherently demonstrate external clinical utility. The clinical relevance of this study lies in its holistic approach to identifying how continuous monitoring can broaden healthcare accessibility while improving precision medicine. Furthermore, it deeply addresses the technical challenges of highly variable ambulatory data quality, the necessity for robust artifact reduction (e.g., via LSTM and GAN architectures), and the limitations of small, homogeneous training datasets. We highlight the essential need for demographic-aware algorithmic models, external validation, and decentralized privacy-preserving models (e.g., federated learning) in diverse populations to ensure the safe, equitable clinical translation of DTx, mHealth, ML, and AI technologies. Full article

The histopathology area is being redefined with the use of artificial intelligence (AI), providing robust tools to help bridge cellular morphology, functional assays, and multi-omics data in stem cell research. The ability of stem cells to undergo self-renewal and differentiation is key in regenerative medicine, but their research requires the careful characterization of morphological and molecular phenotypic traits. Traditional histopathology is invaluable, but its application can be limited by inter-observer variability and restricted scalability. These limitations are circumvented by AI-based techniques, such as machine learning and deep learning, which are capable of classifying cells, performing quantitative morphometry, and forecasting stem cell behavior. Adding AI to genomics, proteomics, and metabolomics will contribute to the further identification of biomarkers and pathways that regulate stem cell fate. This convergence provides new possibilities for precision medicine, personalized therapies, and translational uses like drug discovery and disease modeling. However, its potential has not yet been realized because of the existing difficulties in data quality, variability, regulatory control, and ethical issues, especially in terms of the transparency and justice of AI systems. Emphasized areas for the future include explainable AI, federated learning, and a multimodal framework that integrates imaging, sequencing, and clinical data. Interdisciplinary partnerships and adequate regulatory frameworks will help AI-enabled histopathology reshape stem cell studies and speed up the process of translating regenerative medicine into clinical applications. Full article

Timely and accurate defect classification is critical for ensuring product quality and safety in industrial inspection scenarios. The widespread deployment of Internet of Things (IoT) devices equipped with sensing, computing, and communication capabilities has promoted the development of AI-enabled industrial applications. However, conventional AI approaches typically rely on centralized data collection and processing, which become impractical in real-world IoT environments due to growing privacy concerns and constrained device resources. To address these challenges, this paper proposes a communication-efficient adaptive federated learning algorithm for heterogeneous defect classification tasks. The proposed approach jointly accelerates the training process through three mechanisms: (i) adaptive local updates that balance communication and computation overheads; (ii) parameter compression that trades off communication cost against model accuracy; (iii) joint bandwidth and computation-power allocation that optimizes per-round communication and computation time across participating devices. We further analyze the joint effects of these three mechanisms and provide a convergence analysis. Extensive simulations show that the proposed method achieves competitive classification accuracy while reducing single-round training time by up to 70%. Full article

The rapid advancement of software-defined networking (SDN) has enhanced network programmability, centralized control, and traffic management flexibility, while also increasing exposure to sophisticated attacks targeting the control plane. Although federated learning (FL) enables collaborative intrusion detection without centralized raw data sharing, existing FL-based intrusion detection systems remain vulnerable to plaintext model update leakage, centralized cryptographic trust, limited interpretability, and insufficient validation in operational SDN environments. To address these limitations, this paper presents an explainable federated intrusion detection framework that integrates distributed key generation (DKG), CKKS-based threshold homomorphic encryption, collaborative decryption, and SHapley Additive exPlanations (SHAP). Unlike conventional HE-enabled FL systems that rely on a trusted authority or a globally shared secret key, the proposed framework removes the trusted key-generation dealer, avoids centralized custody of the complete secret key, and prevents any single client or aggregation server from independently decrypting ciphertexts using locally held key material. A gated recurrent unit (GRU)-based model is used for privacy-preserving intrusion detection, and SHAP provides global and local explanations of model decisions. The framework is further deployed in a real-time SDN testbed to evaluate the online inference pipeline following threshold-secured federated training. Computationally intensive cryptographic operations, including DKG, encrypted aggregation, and threshold decryption, are performed during offline training, while the converged global model enables low-latency inference at runtime. Experiments on the InSDN, CICDDoS2017, and CICDDoS2019 datasets with 4, 8, and 12 client federated configurations achieved detection accuracies above 99% across all datasets. The evaluation also examines encryption latency, collaborative decryption overhead, secure aggregation cost, communication complexity, and scalability. The results demonstrate that the proposed framework provides a practical balance among decentralized key management, privacy-preserving aggregation, explainability, detection performance, and real-time SDN deployment feasibility. Full article

Heavy and extra-heavy oils represent a large and growing share of recoverable hydrocarbon resources, yet their extreme viscosity, high heteroatom content, and non-Newtonian behavior routinely defeat empirical correlations developed for conventional crude. Machine learning has emerged as a candidate response to this modeling gap, but existing reviews largely catalog applications without asking whether the technology is actually ready for industrial deployment. This critical review synthesizes machine learning applications across five thematic domains of the heavy-oil value chain: physicochemical property prediction, enhanced oil recovery, flow assurance, reactive recovery, and downstream upgrading. Studies are read through a three-phase historical lens, tracing the field’s progression from empirical-correlation replacement to methodological diversification to physics-informed and closed-loop integration, and evaluated against a Technology Readiness Level (TRL) scale adapted specifically for heavy-oil machine learning. The multilayer perceptron anchors more of the primary corpus than any other architecture, a pattern that, in our interpretation, reflects small-sample, low-dimensional regression needs rather than any demonstrated advantage over other architectures. Enhanced oil recovery is the only cluster to reach organizational-scale deployment, anchored by a single multi-decade operator program, Chevron’s San Joaquin Valley i-field; the remaining clusters are constrained less by modeling sophistication than by single-basin datasets and undisclosed uncertainty. Measured against three falsifiable deployability criteria, fidelity preservation below 10° API, operator-grade interpretability, and demonstrated laboratory-to-field transferability, no study in the reviewed corpus is documented to satisfy all three simultaneously; because industrial implementations are frequently proprietary, this is a statement about the published record identified by this search, not a claim that the capability does not exist. Federated learning, physics-informed architectures, and sequence-aware models emerge as the directions most likely to close this gap. Full article

Federated learning (FL) enables collaborative model training across decentralized clients while preserving data privacy, yet real-world deployments still face communication bottlenecks, performance degradation under heterogeneous client data, and limited personalization. In this study, we introduce FedMCP++, a modular and communication-efficient personalized FL framework in which every client owns a complete private model—a lightweight convolutional backbone with a private expert head—and collaboration is carried out entirely through knowledge exchange rather than parameter exchange. In each round, clients share only temperature-softened class predictions and class-wise feature prototypes computed on a small public proxy set; the server fuses them into an accuracy-weighted teacher and broadcasts the result, and clients realign their models through knowledge distillation, an instance-level contrastive objective, and prototype alignment. We evaluate FedMCP++, its ablations, and two knowledge-based baselines on six benchmark vision datasets with 10, 20, and 30 clients. The results indicate dataset-dependent trade-offs rather than uniform superiority: collaborative distillation improves average client-level accuracy over independent local training in twelve of eighteen configurations—most clearly under severe per-client data scarcity (e.g., up to +2.7 percentage points on KMNIST and +2.4 on STL-10 with 20–30 clients)—whereas independent training ensembles remain strongest on SVHN and CIFAR-10 at the studied budgets. Because no parameters are transmitted, the per-round uplink payload is a fixed-size 42.6 KB message, 9.9–12.8× smaller than full-model synchronization, and is invariant to model capacity. These properties make FedMCP++ a flexible framework for personalized FL in wireless edge and Internet of Things environments where bandwidth and privacy constraints are paramount. Full article

Federated, semi-supervised novelty detection is well suited for intrusion detection on resource-constrained Internet of Things (IoT) nodes: each device learns a model of benign traffic, shares only summary statistics, and does not transmit raw traffic samples. A previously published cross-layer federated detector, Baseline K-Means, showed that periodically merging worker statistics through a coordinator raises the detection rate, but it also exhibited a systematic side effect: after every merge, the precision decays, and the false-positive rate (FPR) climbs because the coordinator recomputes its threshold from streaming distances filtered by the closest observed anomaly, so tightens after every merge, flagging progressively more benign traffic; the threshold was also hand-tuned. We present AF-BKM, an Adaptive Federated Baseline K-Means that repairs the federated mechanism with two label-free, statistics-only enhancements, denoted as E1 and E2: (i) an adaptive decision threshold read from the benign Mahalanobis-distance distribution, requiring no manual percentile search and no attack labels (E1), and (ii) a robust, benignly anchored aggregation that blends worker means under quality weighting and outlier-worker filtering and recalibrates the threshold on a trusted benign anchor to a stable, anchor-referenced false-positive level, which a target-FPR rule can make operator-selectable instead of tightening it toward the nearest anomaly (E2). With MinMax scaling fit only on benign baseline data and non-IID federated streams on NSL-KDD, UNSW-NB15 and the N-BaIoT corpus of real traffic from commercial IoT devices, AF-BKM removes the merge-induced precision decay (the first-to-last-epoch precision change improves from $-0.134$ to $-0.002$ on NSL-KDD, from $-0.121$ to $-0.014$ on UNSW-NB15, and from $-0.170$ to $-0.009$ on N-BaIoT) and reduces the mean FPR by 30–$64\%$, depending on the dataset; all central improvements are significant across 10 seeds (Wilcoxon $p=0.002$, large effect sizes). AF-BKM preserves recall on NSL-KDD and N-BaIoT and, on the harder UNSW-NB15, exposes an explicit precision–recall trade-off through a benign target-FPR knob. In fp32, the deployed model serializes to $5.5$–52 KB, a packet is classified in 11–27 µs on a desktop CPU, and each merge round uploads a $d+3$-value summary (160–472 B) $94.6$–$98.3\%$ smaller than the same summary extended with the covariance upper triangle. A robustness study covering selected faulty-worker updates, contamination of the commissioning anchor, and detector-level white-box evasion reports the measured degradation patterns: fabricated threshold candidates have no direct path to the threshold, although a fabricated mean still reaches it indirectly through the blended centroid, and the anchor-referenced false-positive level remains stable under percent-level anchor contamination, while recall sensitivity is dataset-dependent and the evasion budget tracks the benign–attack margin of each dataset. We frame the contribution with a focused taxonomy that identifies merge-induced precision decay under non-IID workers as an open gap. Code is released for reproducibility. Full article

Predicting customer churn in the telecom sector is critical for retaining subscribers, maintaining brand reputation, and staying ahead of competitors. Losing customers not only reduces revenue but can also weaken long-term market position in a highly competitive industry. While machine learning has been widely used to address this challenge, most traditional approaches depend on centralizing customer data. This raises major concerns about user privacy, data ownership, and compliance with strict regulations such as GDPR. These challenges make it difficult for businesses to fully utilize customer data while safeguarding sensitive information. In this paper, we investigate a privacy-preserving approach to churn prediction that combines federated learning (FL) with differential privacy (DP). Rather than collecting all customer data in a single repository, the investigated framework enables multiple clients to collaboratively train a deep neural network while maintaining data locality during the federated training process. To further enhance privacy protection, we employ Differentially Private Stochastic Gradient Descent (DP-SGD) and add controlled noise to model updates, reducing the possibility of inferring individual data contributions. This work systematically evaluates how different privacy levels, expressed through ε and δ, influence model performance under simulated non-IID client distributions. The experiments analyze the privacy–utility trade-off using multiple evaluation metrics and compare the results with centralized and non-private federated-learning approaches. The findings show that the investigated framework maintains competitive predictive performance across a range of privacy budgets while demonstrating a clear privacy–utility trade-off. Very strict privacy budgets result in substantial performance degradation, particularly for smaller and more imbalanced datasets, whereas moderate privacy budgets maintain competitive predictive performance with limited degradation. This study highlights the potential of privacy-preserving federated learning for practical distributed analytics applications where protecting sensitive data is essential. Full article

Poisoning attacks present a significant challenge for federated learning (FL), particularly when clients operate with Non-IID data and updates are transmitted through multi-tier aggregators. This paper introduces FLVaccin, a hierarchical personalized FL framework structured as an unbalanced tree. In this framework, each non-root node both hosts local clients and aggregates shared MobileNetV2 features (FedPer), whereas the root node does not possess local data. The proposed defense integrates node-level CIFAR-100 vaccination, which calibrates depth- and round-adaptive tolerances, with per-client trend quarantine and root backbone rejection. Experimental results on CIFAR-10 with 100 clients, 25 aggregators, and Dirichlet Non-IID partitioning ($\alpha =0.5$) demonstrate that the clean baseline achieves 79.9% accuracy. In contrast, unconstrained mixed attacks reduce performance to near-chance levels (20.2% k-fold). When vaccination and quarantine are enabled, 535 attack events still result in a 76.5% ± 0.4% k-fold accuracy (77.3% test), remaining within 2.6 percentage points of the clean model. These findings indicate that tree-aware, vaccination-calibrated monitoring can maintain model utility under persistent multi-vector poisoning without the need to share raw data. Full article

With the rapid development of Industrial Internet of Things (IIoT), large amounts of sensor data generated by industrial devices and edge nodes have become the basis of intelligent manufacturing applications. Collaborative modeling on these distributed data is important for tasks such as anomaly detection, equipment monitoring, and predictive maintenance. Federated learning offers a practical way to train models without exposing raw sensor data, but it still faces privacy leakage and malicious poisoning attacks. To address these issues, this paper proposes a hierarchical privacy protection and poisoning-robust defense framework for industrial federated learning. Starting from the sensitivity differences among parameters at different model layers, the proposed method designs a hierarchical privacy-budget allocation strategy that enhances protection for sensitive information while minimizing the performance impact of perturbation. Meanwhile, a multi-layer, multi-feature anomaly-detection mechanism is adopted to identify malicious updates by jointly exploiting directional consistency, scale stability, and inter-layer similarity, and majority voting together with update clipping is used to further improve system robustness. Experiments on Fashion-MNIST, MVTec AD, and C-MAPSS demonstrate that the proposed method can effectively suppress global-model degradation under multiple poisoning attacks and achieves a favorable balance among privacy protection strength, robustness, and training efficiency. Full article

Educational assessment systems have primarily relied on episodic forms of assessment, including examinations, assignments, grades, and credentials. These approaches provide efficient and scalable summaries of achievement and yet capture only part of the developmental process through which learners build competence. Moreover, learning increasingly unfolds across digital platforms, workplaces, collaborative networks, and AI-mediated environments, generating rich evidence of learner development that remains fragmented across systems and contexts. Advances in artificial intelligence, learning analytics, multimodal analytics, learner modeling, and semantic interoperability make it increasingly feasible to connect, integrate, and interpret this evidence across contexts and over time. This paper introduces the AI-Mediated Continuous Assessment Infrastructure (AIM-CAI), a sociotechnical framework supporting longitudinal, probabilistic interpretation of distributed evidence of learning. Within AIM-CAI, continuous assessment refers to the ongoing accumulation and dynamic interpretation of evidence generated through learning activities. The framework integrates distributed evidence systems, evidence serialization mechanisms, AI-mediated semantic translation, probabilistic learner models, dynamic competency profiles, and federated governance architectures to support context-sensitive interpretations of learner development while maintaining human judgment, privacy, accountability, and learner agency. The authors examine implications for assessment, credentialing, lifelong learning, institutional roles, interoperability, and governance and outline a research agenda addressing key psychometric, ethical, and governance challenges, including validity, fairness, surveillance, semantic instability, and ownership of learning evidence. Full article

The transition to a new industrial paradigm beyond Industry 5.0 demands hyperconnectivity and massive processing, creating a paradox where high computational demand can threaten corporate sustainability goals (ESG). This study aims to structure and prioritize a Multilayer Framework of multicloud technologies to balance disruptive advances, lean optimization, and decarbonization. A mixed, sequential, and exploratory-normative approach was used. Initially, the literature was triangulated with expert panels and suppliers to map 21 technological functionalities, structuring them into four layers of a bidirectional value flow. Then, a hybrid multi-criteria modeling (AHP-TOPSIS) was applied to rank these technologies against five market constraints. Calibration with AHP revealed that Cyber Resilience, approximately 38%, and Process Optimization, approximately 27%, lead executive priorities, surpassing environmental impact or cost efficiency. As a result, the TOPSIS ranking highlighted Human–Machine Symbiosis (BCI/neuroergonomic readiness), Zero Trust architecture, Federated Learning, and GenAI KPI Analytics as the leading functionalities in their respective layers, with Hyper-BPM emerging as a closely associated optimization engine at the governance layer. Finally, the proposed roadmap was assessed using an anonymized industrial Proof of Concept (PoC) in a brownfield advanced manufacturing facility, providing evidence of its operational feasibility for integrating lean optimization with legacy systems and ESG-oriented monitoring. It is concluded that industrial sustainability does not rely solely on green technologies, but on decentralized orchestration along the Edge-Cloud continuum. Environmental gains are therefore more likely to emerge when cyber governance and lean-oriented operational management jointly support decentralized multicloud orchestration. Full article