With the increasing need to train AI models on sensitive healthcare data, Federated Learning (FL) has emerged as a decentralized approach that avoids raw data sharing. However, existing methods such as DP-FL and zkFL still suffer from high privacy leakage, computational overhead, and scalability challenges. To overcome these limitations, this study introduces ZK-FedTransformer++, a novel privacy-preserving FL framework. It integrates lightweight TinyViT transformers, zk-SNARKs for verifiable training, differential privacy for statistical protection, and heuristic client selection for robust participation The approach provides secure model updates via cryptographic proof circuits and noise-perturbed gradients. Experiments based on the RSNA Breast Cancer Detection dataset achieve 91.2% accuracy and 35% less privacy leakage. Tools utilized include PyTorch, zk-SNARK libraries, and privacy accounting protocols. In summary, ZK- FedTransformer++ is an effective privacy enhancement, accuracy improvement, and scalability solution that is a feasible solution for secure, decentralized AI applications in real-world healthcare and IoT settings.
Gayatri M Bhandari, Nitin M Shivale, Shrishail S Patil, Pranav Prajapati · 7 authors
Federated learning is an emerging technology that can revolutionize the training of machine learning models. Federated learning refers to an approach to training a machine learning model in a decentralized and collaborative fashion. A central server distributes the model to client devices, where it is trained locally using the clients’ own data. The client then sends the updated model weights to the server, which aggregates them to update the global model. This paper introduces a federated learning platform designed to enable collaborative training of machine learning models across multiple client devices while preserving data privacy. The platform supports a range of supervised learning algorithms, including convolutional neural networks and decision trees, and is compatible with widely used frameworks such as TensorFlow, PyTorch, and Flower. It offers a user-friendly interface where model developers can upload or deploy their machine learning models to a central server. Clients can then access these models and train them locally using their own data. The platform's modular design ensures flexibility in deployment and efficiency in handling real-world applications. The key features of this application include a model repository, secure API access for client integration, local model training capabilities on user-end devices, and a user-friendly UI. The platform aims to democratize machine learning by enabling distributed model training and deployment, promoting collaboration and efficiency across diverse use cases. The scalable infrastructure supports real-time inference, on-device training, and secure data handling, making it ideal for industries ranging from healthcare to finance and beyond.
S N Prajwalasimha, Nilesh Shelke, Dilip Kumar Jang Bahadur Saini, Amit Pimpalkar · 6 authors
Federated Learning (FL) is a decentralized collaborative AI training paradigm that maintains privacy of the data. FL is still susceptible to security attacks, malicious clients, and model integrity issues. To mitigate these issues, we introduce a Blockchain-Enabled Federated Learning (BFL) system that incorporates decentralized ledger technology to provide tamper-evident model aggregation, transparent client engagement, and verifiable updates. The suggested BFL framework uses smart contracts to enable automated trust management, zero-knowledge proofs (ZKPs) to facilitate privacy-enhanced authentication, and an incentive mechanism based on tokenized rewards to promote honest engagement. We also propose an adaptive consensus protocol that maximizes blockchain overhead while preserving high scalability for real-world applications like cybersecurity, healthcare, and Industrial IoT (IIoT). Experimental results on benchmark datasets show that BFL dramatically improves model robustness against data poisoning and adversarial attacks with a 15-25% improvement in attack resilience over state-of-the-art FL methods. Our work presents a complete blueprint for secure, privacy-preserving AI and establishes a foundation for the next generation of decentralized intelligence.
Due to the absence of a trusted model parameter sharing mechanism, model tampering, malicious uploads, and data inconsistency pose significant risks within the Federated Learning (FL) architecture. To address these challenges, this paper proposes an automated Smart Contract-based Asynchronous Model Aggregation (SCAMA) method to enhance model security in decentralized FL frameworks. We design a trigger-based asynchronous model submission mechanism driven by smart contracts. These contracts incorporate a dynamic adjustment strategy that can initiate global aggregation in real time. Furthermore, a decentralized parameter validity verification rule is embedded within the smart contract to ensure that submitted model updates align with the global optimization objective, thereby mitigating risks such as malicious uploads and training drift. Building upon this, we introduce a trustless model parameter-sharing strategy based on on-chain storage. The global model update process is encoded into the smart contract and executed through multiparty verifiable storage on a distributed ledger. During model aggregation, the smart contract utilizes encrypted hashing to verify the consistency of submitted models, ensuring that the parameter versions received by each node are traceable, reliable, and tamper-proof. Experimental results demonstrate that our method offers significant improvements in computational efficiency and model convergence speed compared with traditional synchronous FL approaches and existing asynchronous strategies.
The rising energy demands of large data centers call for energy-efficient AI training methods. Federated Learning (FL), a decentralized paradigm, offers a solution by enabling model training across distributed devices without centralizing sensitive data. This review explores FL's integration with distributed data centers to achieve energy efficiency, analyzing methods like federated averaging and energy-aware protocols to minimize resource use. It highlights techniques such as model compression, quantization, and adaptive FL to reduce on-device computation while maintaining performance. Practical implementation is discussed through tools like TensorFlow Federated and PySyft, with case studies from healthcare, finance, and IoT showcasing cost reductions and sustainability. Future research directions include combining FL with edge computing and low-power AI hardware, emphasizing FL's potential for scalable, sustainable AI.
Multi-Party Computation (MPC), as a distributed computing paradigm, is considered to be a potential solution for providing privacy-preserving for applications following the client-server model. However, traditional MPC solutions cannot satisfy the publicly verifiable requirement of the client-server model. In this paper, we propose a blockchain-based verifiable MPC solution using Pedersen's threshold secret sharing and Lifted ElGamal encryption. We first build a data distribution method using Pedersen's threshold secret sharing and symmetric encryption to protect the privacy of inputs while ensuring robustness. Then, we propose a result processing algorithm using Lifted ElGamal encryption to safeguard the privacy of the outputs. Finally, we employ non-interactive zero-knowledge proof and Pedersen commitment to publicly verify the correctness of the encrypted outputs in the smart contract, enabling the detection of malicious parties. Theoretical analysis indicates that the proposed method can publicly verify the correctness of outputs without revealing plain-text inputs and outputs, which satisfy the privacy-preserving requirements of the client-server model. Experimental evaluations have demonstrated that our proposed approach is efficient regarding computation overhead, communication overhead, and response time in the output verification phase while achieving stronger privacy and robustness.
Federated learning (FL) is an emerging paradigm that enables multiple clients to collaboratively train a machine learning (ML) model without the need to exchange their raw data. However, it relies on a centralized authority to coordinate participants’ activities. This not only interrupts the entire training task in case of a single point of failure, but also lacks an effective regulatory mechanism to prevent malicious behavior. Although blockchain, with its decentralized architecture and data immutability, has significantly advanced the development of FL, it still struggles to withstand poisoning attacks and faces limitations in computational scalability. We propose Zkfhed, a verifiable and scalable FL system that overcomes the limitations of blockchain-based FL in poison attacks and computational scalability. First, we propose a two-stage audit scheme based on zero-knowledge proofs (ZKPs), which verifies that the training data are extracted from trusted organizations and that computations on the data exactly follow the specified training protocols. Second, we propose a homomorphic encryption delegation learning (HEDL), based on fully homomorphic encryption (FHE). It is capable of outsourcing complex computing to external computing resources without sacrificing the client's data privacy. Final, extensive experiments on real-world datasets demonstrate that Zkfhed can effectively identify malicious clients and is highly efficient and scalable in terms of online time and communication efficiency.
One of the goals of Federated Learning (FL) is to collaboratively train a global model using local models from remote participants. However, the FL process is susceptible to various security challenges, including interception and tampering models, information leakage through shared gradients, and privacy breaches that expose participant identities or data, particularly in sensitive domains such as medical environments. Furthermore, the advent of quantum computing poses a critical threat to existing cryptographic protocols through the Shor and Grover algorithms, causing security concerns in the communication of FL systems. To address these challenges, we propose a Post-Quantum Blockchain-based protocol for Federated Learning (PQBFL) that utilizes post-quantum cryptographic (PQC) algorithms and blockchain to enhance model security and participant identity privacy in FL systems. It employs a hybrid communication strategy that combines off-chain and on-chain channels to optimize cost efficiency, improve security, and preserve participant privacy while ensuring accountability for reputation-based authentication in FL systems. The PQBFL specifically addresses the security requirement for the iterative nature of FL, which is a less notable point in the literature. Hence, it leverages ratcheting mechanisms to provide forward secrecy and post-compromise security during all the rounds of the learning process. In conclusion, PQBFL provides a secure and resilient solution for federated learning that is well-suited to the quantum computing era.
Federated Learning (FL) has emerged as a promising paradigm that addresses the delicate balance between data-intensive model development and the preservation of user privacy. Unlike the conventional approach of aggregating large volumes of raw data in a single data center, FL conducts local training on various devices or institutional servers—sometimes referred to as “clients”—and only exchanges model parameters or gradients with a central entity. By design, this decentralized framework keeps personal or proprietary data within the confines of the originating device or organization, significantly reducing the chances of exposing sensitive information. A primary motivation for FL is the ever-increasing concern over privacy violations and compliance with stringent regulations such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA). As global data protection standards continue to evolve, FL offers a compelling solution by minimizing direct data sharing and thereby mitigating the risk of large-scale breaches. Beyond privacy considerations, FL holds practical appeal in many real-world scenarios, including healthcare, finance, the Internet of Things (IoT), and various consumer-focused applications. These sectors routinely handle confidential or regulated data—medical records, bank transactions, or user habits—where a centralized data repository poses both security and compliance hazards. Nevertheless, FL also introduces its own set of challenges. Heterogeneous data distributions across clients can lead to biases and uneven training dynamics. Additionally, new threat vectors—such as model poisoning and inference attacks—have surfaced within decentralized training environments, prompting research into robust security strategies. Furthermore, practical implementation demands careful planning around communication overhead, computational capacity of clients, and the trade-offs that arise when adding privacy guarantees like Differential Privacy or Secure Multi-Party Computation. This paper explores the theoretical underpinnings of Federated Learning, reviews cutting-edge privacy-preserving techniques, examines potential security pitfalls, and presents real-world applications augmented by case studies. We also discuss performance evaluation methods crucial for determining FL’s viability and highlight upcoming research directions that can shape a secure, efficient, and privacy-centered AI ecosystem.
Federated Learning (FL) has emerged as a promising paradigm for secure data sharing in Industrial Internet of Things (IIoT), enabling collaborative model training without direct exchange of raw data. However, recent studies have shown that FL still suffers from privacy vulnerabilities, where adversaries can reconstruct sensitive information by analyzing shared model parameters. Although several privacy-preserving FL (PPFL) schemes have been proposed to address these challenges, they primarily focus on protecting local model privacy, with limited attention to protecting global model confidentiality during aggregation. Additionally, their reliance on centralized aggregation servers introduces risks of single points of failure. To address these challenges, we propose a novel privacy-preserving blockchain-based FL framework (PBFL) that integrates blockchain, homomorphic encryption (HE), and a single masking. Specifically, PBFL employs HE to enable secure model training within the ciphertext domain, ensuring global model confidentiality. The single masking technique allows clients to apply unique random masks to their encrypted local model updates, enabling secure aggregation while preserving local privacy. Additionally, PBFL leverages blockchain for decentralized aggregation and encrypted model storage, effectively mitigating the risks associated with centralized servers. Experimental results demonstrate that PBFL achieves comparable model accuracy to state-of-the-art solutions while providing enhanced privacy protection. Furthermore, even with a client dropout rate of up to 30%, PBFL outperforms other blockchain-based PPFL methods in terms of computational and communication efficiency.
The correctness of a computation can be efficiently verified in a privacy-preserving manner without re-execution using zero-knowledge succinct non-interactive arguments of knowledge (zkSNARKs). With short transcript sizes and fast verification times, zkSNARKs enable the potential deployment of computationally intensive algorithms—such as machine learning models—on the blockchain, making them efficiently verifiable through short proofs. However, the prover time for matrix computations in these settings often fails to scale efficiently with increasing model complexity and data size. We are the first to systematically address zkSNARKs for general matrix computations with practical prover efficiency. We achieve an $O(N + nM)$ prover time, asymptotically faster than the unverified matrix computation, for computations involving $M$ matrix operations on $n \times n$ matrices with $N$ total non-zero entries. Starting with a single dense matrix multiplication, we propose zkMatrix, a special-purpose zkSNARK for verifying committed $n \times n$ matrix multiplication through their projections onto random vectors. Among zkSNARKs with $O(\log n)$ transcript size and verifier time, zkMatrix is the first to achieve $O(n^2)$ prover time and $O(n^2)$ RAM usage. Batching multiple proofs together reduces the prover time for each additional multiplication to $O(n)$ group operations. Next, we design zkSNARKs for sparse matrix multiplication with $N$ non-zero entries. zkSmart reduces the prover time from $O(n^2)$ to $O(N + n)$, relying on an $O(N + n)$-prover vector-matrix-vector product argument, achieved by improving Bulletproofs. Moreover, \zksmart formulates verifiable computation represented as a matrix circuit of $M$ nodes, each denoting a matrix operation. Sparse matrix multiplication translates the matrix circuit satisfiability (Mat-Circ-SAT) problem into the high-dimensional rank-1 constraint system (HD-R1CS), a matrix-circuit version of the rank-1 constraint system (R1CS), traditionally used for arithmetic circuits. Using zkSmart, we achieve $O(N + nM)$ prover time for general matrix computations. To reduce the cost of committing to intermediate variable matrices in zkSmart, we introduce Evalyn, which generates proofs using a pre-order tree traversal on the abstract syntax tree (AST) of a matrix expression. Evalyn ensures output and input consistency in serial matrix computations by linking randomness for zkSNARKs between parent and child nodes, eliminating the need to commit to the nodes and significantly improving prover efficiency. Our prover for R1CS outperforms state-of-the-art general-purpose zkSNARKs. As a foundational component of our framework, we optimize Bulletproofs to construct the fastest known inner product argument (IPA). Additionally, we propose a zero-knowledge transformation that commits to transcript elements with only logarithmic overhead—while maintaining compatibility with post-quantum secure, non-homomorphic commitment schemes. We apply our framework to zero-knowledge machine learning (zkML), providing zkSNARKs for neural networks. We translate floating-point truncations and non-linear activation functions into linear algebra equations that can be verified by our framework. We utilize our framework to generate efficient proofs for the attention layer in large language models (LLMs). After resolving all these challenges, we have thoroughly addressed the design of efficient zkSNARKs for matrix computations.
The rapid advancement of technology and the widespread adoption of various IoT devices have resulted in significant challenges in handling and analyzing the vast and diverse volumes of data they produce. Conventional machine learning methodologies, which require data aggregation in a centralized repository for analysis, are often unfeasible due to the significant data volume, limited communication bandwidth, and strict security and privacy requirements. Federated Learning (FL) presents an innovative resolution by facilitating data analysis to take place directly at the data origin, federating the outcomes to generate results comparable to centralized processing. Due to the advancements in growth of federated learning architecture, a number of open-source frameworks have been established to implement this strategy. These frameworks facilitate a range of applications, including healthcare and finance, as well as IoT and smart city initiatives, by enabling data to remain decentralized while still supporting the development of comprehensive analytical models. This study provides a comprehensive evaluation of these established open-source FL frameworks, examining their applicability across various sectors and highlighting their advantages and limitations. The authors examine the various architectural structures, supported algorithms, level of implementation ease, and community support of each framework. The assessment additionally takes into account the scalability of these frameworks and their flexibility in various data scenarios. Through the assessment of the current capabilities and developmental phases of these FL frameworks, this study aims to support practitioners and researchers in identifying the most suitable tools for their specific needs. This analysis highlights the significance of FL in facilitating secure and effective data analysis in the contemporary technological landscape.
Farhana Javed, Josep Mangues‐Bafalluy, Engin Zeydan, Luis Blanco
This paper proposes a blockchain-enabled framework to enhance trust, transparency, and collaboration in Open Radio Access Network (O-RAN) infrastructures through Federated Learning (FL). Traditional O-RAN architectures and centralized machine learning approaches face challenges when integrating multi-vendor environments, primarily due to lack of trust, proprietary data concerns, and limited interoperability. Our solution transitions from implicit trust, where the reliability of contributions is assumed, to explicit trust, where reputation is verifiably established on-chain. We introduce a blockchain-based reputation mechanism that evaluates the accuracy, integrity, and quality of participants’ model updates within the FL process. Smart contracts automate critical tasks-such as participant registration, model update verification, and reputation scoring-ensuring that data inputs directly influence accountability in a tamper-proof, transparent manner. By deploying the framework on a scalable Layer 2 blockchain (Polygon) testnet and proposing the use of a blockchain oracle within this architectural framework for secure off-chain computations, this work focuses on a conceptual architectural approach by aligning with O-RAN’s architecture to propose and deploy a Decentralized Application (DApp) on the blockchain. The proposed framework emphasizes a conceptual design over performance optimization and is structured to naturally benefit from ongoing improvements in blockchain scalability, which may reduce latency and enhance operational efficiency over time. Smart contracts for crucial processes and reputation calculation are included within our proposed DApp. The implementation of this work is publicly accessiblehttps://github.com/farhanajaved/Reputation_O-RAN.
Xin Wang, Li Jiaqian, Ding Xueshuang, H. Zhang · 5 authors
The problem of data privacy protection in the information age deserves people’s attention. As a distributed machine learning technology, federated learning can effectively solve the problem of privacy security and data silos. Differential privacy(DP) technology is applied in federated learning(FL). By adding noise to raw data and model parameters, it can further enhance the degree of data privacy protection. Over the years, differential privacy technology based on federated learning framework has been developed, which is divided into central differential privacy federated learning(CDPFL) and local differential privacy federated learning(LDPFL). Although differential privacy may reduce the accuracy and convergence of federated learning models while protecting data privacy, researchers have proposed a variety of optimization methods to balance privacy protection and model performance. This paper comprehensively expounds the research status of differential privacy techniques based on the federated learning framework, first providing detailed introductions to federated learning and differential privacy technologies, and then summarizing the development status of two types of federated learning differential privacy(DPFL) techniques respectively; for CDPFL, the paper divides the discussion into first proposal of CDP and typical application examples, the impact of Gaussian mechanisms on model accuracy, optimization based on asynchronous differential privacy, and insights from other scholars; for LDPFL, the paper divides the discussion into first proposal of LDP and typical application examples, processing multidimensional data and improving model accuracy, existing methods and optimization for reducing communication costs, balancing privacy protection and data usability, LDPFL based on the Shuffle model, and insights from other scholars; following this, the paper addresses and summarizes the unique challenges introduced by incorporating differential privacy into federated learning and proposes solutions; finally, based on a summary of existing optimization techniques, the paper outlines future directions and specifically discusses three research ideas for enhancing the optimization effects of federated differential privacy: advanced optimization strategies combining Bayesian methods and the Alternating Direction Method of Multipliers (ADMM), integrating lattice homomorphic encryption techniques from cryptography to achieve more efficient differential privacy protection in federated learning, and exploring the application of zero-knowledge proof techniques in federated learning for privacy protection.
Jianping Yu, Hang Yao, Kai Ouyang, Xiaojun Cao · 5 authors
Federated Learning (FL) enables clients to securely share gradients computed on their local data with the server, thereby eliminating the necessity to directly expose their sensitive local datasets. In traditional FL, the server might take advantage of its dominant position during the model aggregation process to infer sensitive information from the shared gradients of the clients. At the same time, malicious clients may submit forged and malicious gradients during model training. Such behavior not only compromises the integrity of the global model, but also diminishes the usability and reliability of trained models. To effectively address such privacy and security attack issues, this work proposes a Blockchain-based Privacy-preserving and Secure Federated Learning (BPS-FL) scheme, which employs the threshold homomorphic encryption to protect the local gradients of clients. To resist malicious gradient attacks, we design a Byzantine-robust aggregation protocol for BPS-FL to realize the cipher-text level secure model aggregation. Moreover, we use a blockchain as the underlying distributed architecture to record all learning processes, which ensures the immutability and traceability of the data. Our extensive security analysis and numerical evaluation demonstrate that BPS-FL satisfies the privacy requirements and can effectively defend against poisoning attacks.
Existing federated learning (FL) systems are highly vulnerable in terms of security and privacy due to their distributed architecture, facing poisoning attacks and inference attacks from adversaries. Some prior works have combined poisoning defenses with cryptographic tools: Secure Multi-Party Computation, Zero-Knowledge Proof, and Homomorphic Encryption to propose robust secure aggregation methods that provide security and privacy preservation for FL. Recently, Qin et al. (KDD’23) demonstrate that personalized federated learning (pFL) can effectively resist backdoor injection in poisoning attacks. In this paper, we analyze that as the number of malicious attackers increases, pFL remains vulnerable to backdoor attacks. Moreover, we reveal that current robust secure aggregation methods fail to offer efficient and robust backdoor defense for pFL. Therefore, we propose FLIGHT, a robust secure aggregation method for pFL. It implements a lightweight backdoor detection through a two-stage personalized defense mechanism and ensures privacy preservation using communication-efficient two-party secure computation (2PC) protocols. Extensive experiments on diverse datasets and neural networks validate that FLIGHT decreases run-time up to 64× compared by prior work RoFL (S&P’23), and 42× compared to FLAME (USENIX Security’22).
Decentralized Federated Learning improves data privacy and eliminates single points of failure by removing reliance on centralized storage and model aggregation in distributed computing systems. Ensuring the integrity of computations during local model training is a significant challenge, especially before sharing gradient updates from each local client. Current methods for ensuring computation integrity often involve patching local models to implement cryptographic techniques, such as Zero-Knowledge Proofs. However, this approach becomes highly complex and sometimes impractical for large-scale models that use techniques such as random dropouts to improve training convergence. These random dropouts create non-deterministic behavior, making it challenging to verify model updates under deterministic protocols. We propose ProxyZKP, a novel framework combining Zero-Knowledge Proofs with polynomial proxy models to provide computation integrity in local training to address this issue. Each local node combines a private model for online deep learning applications and a proxy model that mediates decentralized model training by exchanging gradient updates. The multivariate polynomial nature of proxy models facilitates the application of Zero-Knowledge Proofs. These proofs verify the computation integrity of updates from each node without disclosing private data. Experimental results indicate that ProxyZKP significantly reduces computational load. Specifically, ProxyZKP achieves proof generation times that are 30-50% faster compared to established methods like zk-SNARKs and Bulletproofs. This improvement is largely due to the high parallelization potential of the univariate polynomial decomposition approach. Additionally, integrating Differential Privacy into the ProxyZKP framework reduces the risk of Gradient Inversion attacks by adding calibrated noise to the gradients, while maintaining competitive model accuracy. The results demonstrate that ProxyZKP is a scalable and efficient solution for ensuring training integrity in decentralized federated learning environments, particularly in scenarios with frequent model updates and the need for strong model scalability.
Leveraging blockchain in Federated Learning (FL) emerges as a new paradigm for secure collaborative learning on Massive Edge Networks (MENs). As the scale of MENs increases, it becomes more difficult to implement and manage a blockchain among edge devices due to complex communication topologies, heterogeneous computation capabilities, and limited storage capacities. Moreover, the lack of a standard metric for blockchain security becomes a significant issue. To address these challenges, we propose a lightweight blockchain for verifiable and scalable FL, namely LiteChain, to provide efficient and secure services in MENs. Specifically, we develop a distributed clustering algorithm to reorganize MENs into a two-level structure to improve communication and computing efficiency under security requirements. Moreover, we introduce a Comprehensive Byzantine Fault Tolerance (CBFT) consensus mechanism and a secure update mechanism to ensure the security of model transactions through LiteChain. Our experiments based on Hyperledger Fabric demonstrate that LiteChain presents the lowest end-to-end latency and on-chain storage overheads across various network scales, outperforming the other two benchmarks. In addition, LiteChain exhibits a high level of robustness against replay and data poisoning attacks.