federated learning

Decentralized Federated Learning with Differential Privacy via Lagrangian Relaxation

Federated learning (FL) offers a promising paradigm for training machine learning models on decentralized data sources without directly exchanging data. However, achieving robust privacy guarantees alongside high model accuracy remains a significant challenge. This paper introduces a novel approach to decentralized federated learning leveraging Lagrangian relaxation and differential privacy. Our method allows for a more nuanced and adaptive trade-off between privacy and accuracy by optimizing a Lagrangian function at each participating node. This decentralized optimization process enables a flexible solution to the inherent tension between privacy preservation and model performance. We demonstrate the effectiveness of this approach through a theoretical analysis and outline a potential implementation strategy. The key innovation lies in the localized optimization of the Lagrangian, mitigating the communication bottlenecks often encountered in centralized FL and offering a pathway towards truly decentralized privacy-preserving learning.

A Device-Agnostic Framework for Personalized Anomaly Detection in Wearable and Mobile Ecosystems: Applications in Personal Safety and Respiratory Health Surveillance

Abstract—This paper presents a reformulated, device-agnostic architectural framework for two interrelated applications: (i) personalized audio-based personal safety threat detection, and (ii) respiratory pattern anomaly detection for early-stage epidemiological screening. Originating from an ideation dialogue, the proposed concepts are critically re-examined against fundamental constraints in machine learning, embedded systems, privacy engineering, and sensor heterogeneity. We propose a generalized abstraction layer that decouples sensing modality from inference logic, enabling deployment across heterogeneous smart devices (wearables, smartphones, IoT nodes) without hardware-specific dependencies. We address the zero-positive-example training problem through a reformulation as one-class classification and temporal anomaly detection. Privacy is preserved via on-device federated feature extraction with no raw data transmission. We further analyze the practical limitations identified by domain experts and propose mitigations grounded in current literature. The framework is positioned as a foundational architecture rather than a deployable product, inviting interdisciplinary validation. Impact Statement—By introducing a sensing abstraction layer and reformulating personal safety and respiratory monitoring as one-class anomaly detection problems, this work provides a portable, privacy-preserving architecture that can operate across heterogeneous consumer devices without hardware-specific redesign. The framework addresses the zero-positive-example constraint inherent to rare-event detection and offers a structured research agenda for empirical validation, potentially accelerating the development of reliable, edge-deployed health and safety monitoring systems. Index Terms—Anomaly detection, one-class classification, device-agnostic computing, respiratory pattern analysis, personal safety systems, federated learning, wearable computing, edge inference.

Meta-Learning for Adaptive Model Selection in Federated Learning

Federated Learning (FL) has emerged as a promising approach for training machine learning models across decentralized devices while preserving data privacy. However, a significant limitation of traditional FL is the static nature of model selection. Typically, all clients are trained with the same base model, regardless of the heterogeneity in their local data distributions. This static approach often leads to suboptimal performance, particularly in scenarios with diverse client datasets. This paper proposes a novel meta-learning framework to address this limitation. The framework learns a central model that dynamically selects the most appropriate base model for each client based on its local data characteristics. This adaptive model selection process aims to improve the overall system performance and efficiency of federated learning. The key contributions are the introduction of adaptability into the model selection stage and the development of a meta-learning approach for achieving this adaptation.

Digital Twins for Sustainable Groundwater Resources Management: From Monitoring and Prediction to Governance and Resilience—A Review

This article presents a scoping review of the literature on digital twins (DTs) for sustainable groundwater resources management, which constitutes a very recent and rapidly expanding research field, with literature moving quickly from conceptual frameworks to application-oriented systems. The literature, selected through the PRISMA 2020 methodology, is organized into seven sub-topics: AI and ML foundations, digital twin architectures and frameworks, aquifer-scale DT applications, agricultural and water–energy–food (WEF) nexus DTs, basin and urban water DTs, sensing, monitoring and IoT infrastructures, and governance, resilience and socio-hydrology. This structure shows how the field is shifting from monitoring and prediction toward integrated decision support, where process-based models, machine learning surrogates, real-time sensing and optimization are combined to support drought mitigation, saltwater intrusion control, irrigation management, climate adaptation and basin-scale planning. Across the reviewed studies, the most recurrent contributions are the construction of hybrid model architectures, the use of DTs to close the loop between observation and control, and the growing recognition that groundwater management must incorporate governance, stakeholder decision-making and socio-hydrological feedbacks. At the same time, the literature still faces key limitations, especially uncertainty quantification, interoperability between models and data streams, transferability to data-scarce settings and limited validation under real operational conditions. Future research should therefore focus on physics-informed and explainable AI, federated and scalable DT architectures, stronger coupling with socio-hydrological and governance frameworks, and more field-tested implementations that can demonstrate robust performance across diverse hydrogeological and institutional contexts.

Decentralized Learning with Federated Bayesian Networks

This paper proposes a novel decentralized learning algorithm for Bayesian networks, termed Federated Bayesian Networks (FBNs). The core idea is to enable nodes within a network to learn independently and collaboratively, mirroring the principles of federated learning. Each node maintains its own Bayesian network and updates its parameters based on probabilistic information received from its neighbors. This approach avoids the need for centralized data aggregation, addressing key challenges associated with privacy and scalability in traditional Bayesian network learning. The algorithm iteratively refines both the network structure and its parameters, leading to a more accurate and robust global model. Mathematical formulations are presented to detail the update rules and convergence properties of the FBN algorithm. The key contribution lies in establishing a framework for distributed Bayesian network learning, particularly well-suited for scenarios with heterogeneous data and limited communication bandwidth. This work lays the foundation for applying FBNs to diverse applications, including healthcare, smart cities, and anomaly detection.

Blockchain-Based Federated Learning with Differential Privacy

Federated learning (FL) offers a promising approach to training machine learning models on decentralized data sources without direct data sharing. However, traditional FL systems remain vulnerable to privacy breaches and data manipulation. This paper proposes a novel blockchain-based architecture that addresses these concerns by integrating data provenance tracking, integrity verification, and differential privacy mechanisms. The system leverages blockchain technology to create an immutable record of model updates and data contributions, ensuring transparency and accountability. Simultaneously, differential privacy techniques are applied during the training process to protect the privacy of individual data contributors. This combined approach significantly enhances the security and trustworthiness of FL systems, enabling secure and collaborative model training across diverse data sources. The proposed system utilizes cryptographic hashing and Merkle trees to guarantee data integrity and employs noise injection strategies within differential privacy mechanisms to protect user data. The core contribution is a secure and verifiable FL framework with enhanced privacy guarantees.

DIGITAL TWINS AND FEDERATED LEARNING IN CARDIOMETABOLIC HEALTH: A SYSTEMATIC REVIEW AND A NOVEL PUBLIC HEALTH FRAMEWORK FOR OBESITY, TYPE 2 DIABETES, AND CARDIOVASCULAR DISEASE INTEGRATION

Meta-Learning for Adaptive Model Selection in Federated Learning

Federated Learning (FL) has emerged as a promising approach for training machine learning models across decentralized devices while preserving data privacy. However, a significant limitation of traditional FL is the static nature of model selection. Typically, all clients are trained with the same base model, regardless of the heterogeneity in their local data distributions. This static approach often leads to suboptimal performance, particularly in scenarios with diverse client datasets. This paper proposes a novel meta-learning framework to address this limitation. The framework learns a central model that dynamically selects the most appropriate base model for each client based on its local data characteristics. This adaptive model selection process aims to improve the overall system performance and efficiency of federated learning. The key contributions are the introduction of adaptability into the model selection stage and the development of a meta-learning approach for achieving this adaptation.

Decentralized Federated Learning with Differential Privacy for Scientific Data

This paper presents a novel approach to collaborative scientific data analysis leveraging Decentralized Federated Learning with Differential Privacy (DFLDP). The core challenge in many scientific domains is the reluctance to share raw data due to stringent privacy regulations and intellectual property protections. Traditional Federated Learning (FL) solutions, while offering a degree of data privacy, still rely on centralized aggregation, a point of vulnerability. Our proposed DFLDP framework addresses this limitation by adopting a decentralized architecture where individual researchers maintain complete control over their datasets. Crucially, we integrate differential privacy mechanisms directly into the aggregation process, adding a quantifiable layer of protection against data leakage. This ensures that the learned model benefits from the collective knowledge of multiple researchers without revealing individual data contributions. The system utilizes a gossip-based communication protocol for model updates, minimizing communication overhead. We formally define the mathematical framework, outlining the key components and their interactions. The system's performance is evaluated in a simulated environment, demonstrating the effectiveness of the DFLDP approach in achieving accurate models while upholding stringent privacy guarantees. The core claim of this work is that sharing raw scientific data for federated learning is often prohibited due to privacy concerns and intellectual property restrictions. The core mechanism implemented is the realization of a federated learning system that utilizes differential privacy to protect data during aggregation, while also employing a decentralized architecture where individual researchers retain control over their data. This new approach combines federated learning with differential privacy and decentralization, enabling collaborative scientific discovery without compromising data privacy or intellectual property rights.

Decentralized Federated Learning with Byzantine Fault Tolerance using Blockchain-Based Verification

Federated learning (FL) offers a promising approach to training machine learning models on decentralized datasets without directly sharing the data itself. However, existing FL systems are susceptible to Byzantine attacks, where malicious participants can inject poisoned data or manipulate model updates, ultimately compromising the global model's integrity. This paper proposes a novel decentralized federated learning system incorporating Byzantine fault tolerance (BFT) achieved through a blockchain-based verification layer. The system leverages blockchain technology to cryptographically verify model updates from each participant before they are aggregated into the global model. This ensures data integrity and enables the detection and mitigation of Byzantine attacks. The core claim is that existing FL systems are vulnerable to these attacks. The proposed mechanism implements a decentralized FL system utilizing a blockchain-based verification layer. This represents a new approach to robust FL training. The system utilizes the following key components: participant nodes, a blockchain network, and a global model aggregator. The blockchain network is composed of multiple nodes that maintain a ledger of all model updates. The global model aggregator utilizes the blockchain network to verify model updates before aggregating them into the global model. The proposed system enhances the security and robustness of FL training by providing a tamper-proof audit trail and enabling the detection of malicious participants.

A Systematic Literature Review of Emerging Big Data Analytics Using Hadoop: Applications, Trends, and Implications for Sustainability

The rapid expansion of Big Data has transformed data-driven decision-making across diverse sectors, including healthcare, cybersecurity, smart cities, and the Internet of Things (IoT). However, traditional data management platforms struggle to address the increasing volume, velocity, variety, and veracity of modern data ecosystems, resulting in scalability constraints and inefficiencies. These challenges necessitate distributed, cloud-native, and energy-efficient architectures capable of supporting Artificial Intelligence (AI) and Machine Learning (ML)–driven analytics. This paper presents a systematic literature review (SLR) of emerging Big Data analytics frameworks centered on the Hadoop ecosystem and its modern extensions, including Apache Spark, Flink, Kafka, and data lakehouse technologies. Following the PRISMA 2020 guidelines, the review analyzes 138 primary studies published between 2015 and 2025, with 60 studies selected through rigorous inclusion and quality assessment criteria. The analysis evaluates architectural evolution from batch-oriented MapReduce systems to unified batch-stream processing and cloud-native data platforms, emphasizing performance optimization, resource efficiency, and sustainable computing practices. The findings reveal significant progress in integrating AI and ML pipelines with Hadoop-based infrastructures for applications such as real-time fraud detection, anomaly detection, predictive healthcare analytics, renewable energy forecasting, and intelligent urban management. Particular attention is given to explainable AI (XAI), federated data processing, and green computing strategies that enhance transparency, trust, and environmental sustainability. This review proposes a structured taxonomy of Hadoop-based Big Data systems, categorizes emerging research trends, benchmarks modern frameworks, and identifies open research challenges. The study provides theoretical, practical, and policy-level implications to guide researchers, system architects, and decision-makers in selecting scalable, low-latency, and energy-aware Big Data solutions aligned with next-generation digital transformation and sustainable development objectives. Not applicable, as this study is not a clinical trial.

Spatiotemporal vision transformers with Byzantine-robust federated prompt tuning for continuous urban perception

Introduction While spatial Vision Transformers (ViTs) achieve high precision in urban scene parsing, their frame-by-frame application in autonomous driving suffers from severe temporal flickering and prohibitive retraining costs across decentralized vehicle fleets. Methods To overcome these dual bottlenecks, this paper introduces a unified Spatiotemporal Hierarchical Mask-Refinement (ST-HMR) framework integrated with a Byzantine-Robust Federated Learning (BR-FL) protocol. The ST-HMR module caches fine-grained prompt tokens via an asymmetrical Temporal Cross- Attention buffer to enforce inter-frame geometric continuity. Concurrently, the BR-FL pipeline employs Multi-Krum geometric distance filtration to aggregate 3 localized prompt gradients from decentralized fleets securely, updating only a 1.4% active parameter subset. Results Evaluated on the Cityscapes Video dataset, the ST-HMR framework improves the video segmentation mean Intersection over Union (mIoU) to 83.5%, elevates the Temporal Consistency (TC) score to 88.5, and reduces depth Absolute Relative Error (Abs Rel) to 0.085, all while maintaining real-time edge processing at 38 FPS. Under severe adversarial network conditions (up to 30% Byzantine/malicious sensor nodes), the BR-FL protocol achieves a 98.4% Byzantine detection rate and maintains a global mIoU of 81.9%, while reducing Over-The-Air (OTA) transmission payloads by over 99% (3.8 MB vs. 1.2 GB per round). Discussion These findings demonstrate that parameter-efficient prompt caching eliminates temporal boundary jitter without heavy 3D transformer overhead, while geometric gradient filtering provides robust defense against decentralized poisoning, establishing a scalable, secure, and temporally coherent perception paradigm for next-generation edge robotics.