EnFed enables resource-constrained devices to obtain accurate human activity recognition models via incentive-driven aggregation of nearby device updates, claiming over 95% accuracy and over 90% lower response time than cloud-only frameworks.
Lightfed: An efficient and secure federated edge learning system on model splitting,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2024 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
EnFed: An Energy-aware Federated Learning in Resource Constrained Environments for Human Activity Recognition
EnFed enables resource-constrained devices to obtain accurate human activity recognition models via incentive-driven aggregation of nearby device updates, claiming over 95% accuracy and over 90% lower response time than cloud-only frameworks.