![[2602.14051] Decentralized Federated Learning With Energy Harvesting Devices](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.14051] Decentralized Federated Learning With Energy Harvesting Devices
Summary
The paper explores decentralized federated learning (DFL) using energy harvesting devices, addressing battery depletion issues and proposing a decentralized algorithm for improved efficiency.
Why It Matters
As the demand for machine learning on edge devices grows, this research highlights a sustainable approach to federated learning. By integrating energy harvesting, it enhances the longevity and performance of devices in decentralized networks, which is crucial for real-world applications in IoT and mobile computing.
Key Takeaways
- Decentralized federated learning can be enhanced by energy harvesting techniques.
- The proposed algorithm reduces communication overhead and computational complexity.
- Theoretical analysis shows the algorithm achieves asymptotic optimality.
- Numerical experiments validate the effectiveness of the proposed approach.
- Energy supply impacts convergence in decentralized learning systems.
Computer Science > Machine Learning arXiv:2602.14051 (cs) [Submitted on 15 Feb 2026] Title:Decentralized Federated Learning With Energy Harvesting Devices Authors:Kai Zhang, Xuanyu Cao, Khaled B. Letaief View a PDF of the paper titled Decentralized Federated Learning With Energy Harvesting Devices, by Kai Zhang and 2 other authors View PDF HTML (experimental) Abstract:Decentralized federated learning (DFL) enables edge devices to collaboratively train models through local training and fully decentralized device-to-device (D2D) model exchanges. However, these energy-intensive operations often rapidly deplete limited device batteries, reducing their operational lifetime and degrading the learning performance. To address this limitation, we apply energy harvesting technique to DFL systems, allowing edge devices to extract ambient energy and operate sustainably. We first derive the convergence bound for wireless DFL with energy harvesting, showing that the convergence is influenced by partial device participation and transmission packet drops, both of which further depend on the available energy supply. To accelerate convergence, we formulate a joint device scheduling and power control problem and model it as a multi-agent Markov decision process (MDP). Traditional MDP algorithms (e.g., value or policy iteration) require a centralized coordinator with access to all device states and exhibit exponential complexity in the number of devices, making them impractical for large-scal...
