HAR benchmark datasets

The 'HAR benchmark datasets' contain data used to evaluate Human Activity Recognition systems, focusing on the effectiveness of various techniques in mitigating label leakage during federated learning.

Papers using HAR benchmark datasets (6)

PQS-BFL: A Post-Quantum Secure Blockchain-based Federated Learning Framework2025Label Leakage in Federated Inertial-based Human Activity Recognition2025Fed-Focal Loss for imbalanced data classification in Federated Learning2020 · 34 citesFLAME: Federated Learning Across Multi-device Environments2022 · 3 citesFedSiKD: Clients Similarity and Knowledge Distillation: Addressing Non-i.i.d. and Constraints in Federated Learning2024 · 1 citesTowards Adaptive Asynchronous Federated Learning for Human Activity Recognition2024