Blockchained On-Device Federated Learning

By Hyesung Kim, Jihong Park, Medhi Bennis, Seong-Lyun Kim. 2018

737 citations recorded by the source corpus.

By leveraging blockchain, this letter proposes a blockchained federated learning (BlockFL) architecture where local learning model updates are exchanged and verified. This enables on-device machine learning without any centralized training data or coordination by utilizing a consensus mechanism in blockchain. Moreover, we analyze an end-to-end latency model of BlockFL and characterize the optimal block generation rate by considering communication, computation, and consensus delays.

Subfields: Artificial Intelligence, Computer Networks and Communications

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Networks and Communications

Citations in the corpus

Incentive Design for Efficient Federated Learning in Mobile Networks: A Contract Theory Approach (2019, background): “However, the most existing work made an optimistic assumption that all the mobile devices will unconditionally participate in federated learning when invited [8], [9], which is not practical in the real world due to resource costs incurred by model training [10].”
Wireless Network Intelligence at the Edge (2018, methodology): “On another level, the training process can become robust against malicious and/or malfunctioning devices by the aid of blockchain technologies [172], as elaborated next.”
The Roadmap to 6G Security and Privacy (2021, background): “Since data is the fuel for AI algorithms, it is crucial to ensure their integrity and their provenance from trusted sources [74].”
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection (2019, methodology): “ion Environment (TEE) environments. Ghosh et al. [172] considers the model robustness upon Byzantine parties (or abnormal and adversarial parties). Another potential approach can be blockchain [173], [174]. Preuveneers et al. [173] proposed a permissioned blockchain-based federated learning method t”
Reliable Federated Learning for Mobile Networks (2019, background): “[13].”
Communication-Efficient and Distributed Learning Over Wireless Networks: Principles and Applications (2020, background): “Furthermore, it is vulnerable to a single point of attack or failure [63].”
Decentralized Federated Learning: A Survey and Perspective (2023, methodology): “Kim et al. [110] and Qu et al. [111] have explored using blockchain technology to mitigate SPoF by replacing the central server role, but this approach diverges from the conventional CFL model.”
Federated Learning for Internet of Things: Recent Advances, Taxonomy, and Open Challenges (2020, background): “On the other hand, the local learning model updates of all the end-devices are transferred between miner-enabled base stations for a blockchain-based federated learning model [74].”
Federated Learning With Blockchain for Autonomous Vehicles: Analysis and Design Challenges (2020, methodology): “There are few more recent works which focus on the integration and decentralization of the trust management and reputation management mechanisms using blockchain-based IoT applications [16], [19], [40], [41].”
Federated Learning Meets Blockchain in Edge Computing: Opportunities and Challenges (2021, methodology): “First, we recall the key concept of the classical FL which has been widely ... used in distributed AI systems.”
Convergence Time Optimization for Federated Learning Over Wireless Networks (2020, background): “The authors in [5] proposed a blockchain-based FL scheme that enables edge devices to train FL without sending FL model parameters to a central controller.”
A Survey on Federated Learning for Resource-Constrained IoT Devices (2021, neutral): “Some blockchain paradigm for on-device training is discussed in [36] and [222], but they designed that the framework without considering the challenges of the weak-processing unit and limited memory of IoT devices.”
Communication and Consensus Co-Design for Low-Latency and Reliable Industrial IoT Systems (2019, methodology): “In terms of the scalability, Blockchain is one of the most popular and reliable consensus protocols, utilized in many applications ranging from cypto-currency [9] to distributed machine learning [10] and drone-aided mobile edge computing [11].”
Federated Learning for Healthcare Informatics (2019, background): “-designed incentives, self-interested mobile devices will be reluctant to participate in federal learning tasks, which will hinder the adoption of federated learning [65]. For this reason, Kim et al. [68] introduced the reward mechanism which is proportional to the training sample sizes into the pro”