Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learn More by Using Less: Distributed Learning with Energy-Constrained Devices

Roberto Pereira, Cristian J. Vaca-Rubio, Luis Blanco

TL;DR

This work tackles energy constraints in distributed Federated Learning by introducing LeanFed, an energy-aware framework that reduces each device's local data usage to preserve battery life while keeping participation across many communication rounds. The method defines a per-device data fraction η_e(λ) based on energy budgets B_e, per-round costs b̄_e, and a target round count R, guaranteeing device availability and enabling a near-FedAvg aggregation when η_e = 1. Through experiments on CIFAR-10 and CIFAR-100 with varying data heterogeneity, LeanFed consistently improves final accuracy and stability over the FedAvg baseline, especially under high heterogeneity and limited battery. The results highlight the practicality of energy-efficient, privacy-preserving FL for resource-constrained networks, with potential impact on scalable pervasive AI.

Abstract

Federated Learning (FL) has emerged as a solution for distributed model training across decentralized, privacy-preserving devices, but the different energy capacities of participating devices (system heterogeneity) constrain real-world implementations. These energy limitations not only reduce model accuracy but also increase dropout rates, impacting on convergence in practical FL deployments. In this work, we propose LeanFed, an energy-aware FL framework designed to optimize client selection and training workloads on battery-constrained devices. LeanFed leverages adaptive data usage by dynamically adjusting the fraction of local data each device utilizes during training, thereby maximizing device participation across communication rounds while ensuring they do not run out of battery during the process. We rigorously evaluate LeanFed against traditional FedAvg on CIFAR-10 and CIFAR-100 datasets, simulating various levels of data heterogeneity and device participation rates. Results show that LeanFed consistently enhances model accuracy and stability, particularly in settings with high data heterogeneity and limited battery life, by mitigating client dropout and extending device availability. This approach demonstrates the potential of energy-efficient, privacy-preserving FL in real-world, large-scale applications, setting a foundation for robust and sustainable pervasive AI on resource-constrained networks.

Learn More by Using Less: Distributed Learning with Energy-Constrained Devices

TL;DR

This work tackles energy constraints in distributed Federated Learning by introducing LeanFed, an energy-aware framework that reduces each device's local data usage to preserve battery life while keeping participation across many communication rounds. The method defines a per-device data fraction η_e(λ) based on energy budgets B_e, per-round costs b̄_e, and a target round count R, guaranteeing device availability and enabling a near-FedAvg aggregation when η_e = 1. Through experiments on CIFAR-10 and CIFAR-100 with varying data heterogeneity, LeanFed consistently improves final accuracy and stability over the FedAvg baseline, especially under high heterogeneity and limited battery. The results highlight the practicality of energy-efficient, privacy-preserving FL for resource-constrained networks, with potential impact on scalable pervasive AI.

Abstract

Federated Learning (FL) has emerged as a solution for distributed model training across decentralized, privacy-preserving devices, but the different energy capacities of participating devices (system heterogeneity) constrain real-world implementations. These energy limitations not only reduce model accuracy but also increase dropout rates, impacting on convergence in practical FL deployments. In this work, we propose LeanFed, an energy-aware FL framework designed to optimize client selection and training workloads on battery-constrained devices. LeanFed leverages adaptive data usage by dynamically adjusting the fraction of local data each device utilizes during training, thereby maximizing device participation across communication rounds while ensuring they do not run out of battery during the process. We rigorously evaluate LeanFed against traditional FedAvg on CIFAR-10 and CIFAR-100 datasets, simulating various levels of data heterogeneity and device participation rates. Results show that LeanFed consistently enhances model accuracy and stability, particularly in settings with high data heterogeneity and limited battery life, by mitigating client dropout and extending device availability. This approach demonstrates the potential of energy-efficient, privacy-preserving FL in real-world, large-scale applications, setting a foundation for robust and sustainable pervasive AI on resource-constrained networks.

Paper Structure

This paper contains 8 sections, 4 equations, 3 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Motivation
  3. Energy-aware Federated Learning
  4. Numerical Evaluation
  5. Experimental Setup
  6. Learning on Battery Constrained Devices
  7. Limitations
  8. Conclusion

Figures (3)

  • Figure 1: Test accuracy (y-axis) over communication rounds (x-axis) for the baseline FedAvg (solid lines) and LeanFed (dashed lines) across various levels of data heterogeneity (indicated by color) and full device participation. Top row shows results for the CIFAR-10 dataset, while the bottom row presents results for the CIFAR-100.
  • Figure 2: Number of communication rounds after which devices become inactive due to drained battery, considering $50$ devices, $R = 200$ and $\gamma = 0.5$. Each boxplot corresponds to a different method and/or participation method.
  • Figure 3: Test accuracy of the FedAvg (solid line) and LeaFed (dashed line) for fixed $\gamma=0.5$ on CIFAR-10 (top) and different participation rates (different colors).