Crossing Domain Borders with Federated Few-Shot Adaptation

Manuel Röder, Maximilian Munch, Christoph Raab, Frank Michael Schleif

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

Federated Learning has gained significant attention as a data protecting paradigm for decentralized, clientside learning in the era of interconnected, sensor-equipped edge devices. However, practical applications of Federated Learning face three major challenges: First, the expensive data labeling process required for target adaptation involves human participation. Second, the data collection process on client devices suffers from covariate shift due to environmental impact on attached sensors, leading to a discrepancy between source and target samples. Third, in resource-limited environments, both continuous or regular model updates are often infeasible due to limited data transmission capabilities or technical constraints on channel availability and energy efficiency. To address these challenges, we propose FedAcross, an efficient and scalable Federated Learning framework designed specifically for real-world client adaptation in industrial environments. It is based on a pre-trained source model that includes a deep backbone, an adaptation module, and a classifier running on a powerful server. By freezing the backbone and the classifier during client adaptation on resourceconstrained devices, we enable the domain adaptive linear layer to solely handle target domain adaptation and minimize the overall computational overhead. Our extensive experimental results validate the effectiveness of FedAcross in achieving competitive adaptation on low-end client devices with limited target samples, effectively addressing the challenge of domain shift. Our framework effectively handles sporadic model updates within resource-limited environments, ensuring practical and seamless deployment.

Original languageEnglish
Title of host publicationProceedings of the 13th International Conference on Pattern Recognition Applications and Methods
EditorsModesto Castrillon-Santana, Maria De Marsico, Ana Fred
PublisherSciTePress
Pages511-521
Number of pages11
ISBN (Print)9789897586842
DOIs
Publication statusPublished - 2024
Event13th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2024 - Rome, Italy
Duration: 24-Feb-202426-Feb-2024

Publication series

NameInternational Conference on Pattern Recognition Applications and Methods
Volume1
ISSN (Electronic)2184-4313

Conference

Conference13th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2024
Country/TerritoryItaly
CityRome
Period24/02/202426/02/2024

Keywords

  • Deep Transfer Learning
  • Domain Adaptation
  • Federated Learning
  • Few-Shot Learning
  • Resource Constraints
  • Sporadic Model Updates

Fingerprint

Dive into the research topics of 'Crossing Domain Borders with Federated Few-Shot Adaptation'. Together they form a unique fingerprint.

Cite this