LightTune: Lightweight Forward-Only Online Fine-Tuning with Applications to Link Adaptation

TL;DR

LightTune is a lightweight, backpropagation-free online fine-tuning framework that adapts ML models in real-time on resource-constrained devices, demonstrated on 6G mobile systems for improved link adaptation.

Contribution

We introduce LightTune, a novel online fine-tuning method that is computationally efficient and provably convergent, suitable for deployment on mobile devices.

Findings

Reduced BLER prediction error by up to 48.8% with LightTune.

Achieved up to 15.5% throughput improvement in 6G link adaptation.

Demonstrated real-time adaptation to unseen channel conditions.

Abstract

Deploying machine learning (ML) algorithms on mobile phones is bottlenecked by performance degradation under dynamic, real-world conditions that differ from the offline training conditions. While continual learning and adaptation are essential to mitigate this distributional shift, conventional online learning methods are often computationally prohibitive for resource-constrained devices. In this paper, we propose LightTune, a lightweight, backpropagation-free online fine-tuning framework with provable convergence guarantees. LightTune opportunistically refines ML models using live test-time data only when performance falls below a predefined threshold, ensuring minimal computational overhead and highly efficient responsiveness. As a practical demonstration, we integrate LightTune into a block error rate (BLER) prediction algorithm for 6G mobile systems. This integration enables the ML BLER prediction model to dynamically adapt to previously unseen channel conditions in real-time. Our extensive results show a substantial reduction in the average BLER prediction error of up to 48.8% with online fine-tuning. Furthermore, we leverage this BLER prediction algorithm for link adaptation and demonstrate average throughput improvements of up to 15.5% compared to a conventional table-based outer loop link adaptation (OLLA) algorithm.