Multi-Frequency Federated Learning for Human Activity Recognition Using Head-Worn Sensors

2024-06-17

Australasian Conference on Interactive Entertainment

AI Summary

To address privacy risks arising from centralized data collection and the challenge of federated learning coordination across heterogeneous sampling frequencies in head-mounted sensor-based human activity recognition (HAR), this paper proposes MF-FL-HAR—the first multi-frequency federated learning framework tailored for ear-worn and other head-mounted devices. MF-FL-HAR integrates a frequency alignment mechanism, a multi-branch neural network architecture, and hierarchical federated optimization to enable privacy-preserving collaborative modeling under heterogeneous sampling rates (e.g., 50/100/200 Hz). Evaluated on two real-world head-mounted HAR datasets, MF-FL-HAR consistently outperforms single-frequency baselines, achieving average accuracy improvements of 3.2–5.7%. This work is the first to empirically validate the feasibility and effectiveness of cross-frequency federated HAR. The implementation code is publicly available.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Security and Privacy: Large-scale security measurements

Abstract

Human Activity Recognition (HAR) benefits various application domains, including health and elderly care. Traditional HAR involves constructing pipelines reliant on centralized user data, which can pose privacy concerns as they necessitate the uploading of user data to a centralized server. This work proposes multi-frequency Federated Learning (FL) to enable: (1) privacy-aware ML; (2) joint ML model learning across devices with varying sampling frequency. We focus on head-worn devices (e.g., earbuds and smart glasses), a relatively unexplored domain compared to traditional smartwatch- or smartphone-based HAR. Results have shown improvements on two datasets against frequency-specific approaches, indicating a promising future in the multi-frequency FL-HAR task. The proposed network’s implementation is publicly available for further research and development.**

Problem

• Privacy-aware machine learning for human activity recognition
• Joint model learning across devices with varying sampling frequencies
• Focus on head-worn sensors, an underexplored domain in HAR

Innovation

• Multi-frequency Federated Learning for privacy-aware ML
• Joint model learning across devices with varying sampling rates
• Focus on head-worn sensors like earbuds and smart glasses