Federated Reinforcement Learning for Runtime Optimization of AI Applications in Smart Eyewears

TL;DR

This paper introduces a federated reinforcement learning framework for smart eyewear devices, enabling collaborative AI model training that enhances stability and privacy despite limited device resources and network constraints.

Contribution

It presents a novel FRL approach with synchronous and asynchronous strategies tailored for real-time AI in resource-constrained smart eyewear devices.

Findings

Federated agents show reduced performance variability.

FRL improves stability and reliability of AI applications.

Framework supports real-time object detection in SEWs.

Abstract

Extended reality technologies are transforming fields such as healthcare, entertainment, and education, with Smart Eye-Wears (SEWs) and Artificial Intelligence (AI) playing a crucial role. However, SEWs face inherent limitations in computational power, memory, and battery life, while offloading computations to external servers is constrained by network conditions and server workload variability. To address these challenges, we propose a Federated Reinforcement Learning (FRL) framework, enabling multiple agents to train collaboratively while preserving data privacy. We implemented synchronous and asynchronous federation strategies, where models are aggregated either at fixed intervals or dynamically based on agent progress. Experimental results show that federated agents exhibit significantly lower performance variability, ensuring greater stability and reliability. These findings underscore the potential of FRL for applications requiring robust real-time AI processing, such as real-time object detection in SEWs.