Domain Adaptation Under Wireless Network Constraints: When Does It Become Green?
2026-06-22 • Artificial Intelligence
Artificial IntelligenceMachine Learning
AI summaryⓘ
The authors study how changing conditions in 6G wireless networks make it hard for data models to keep working well over time. They look at a method called Unsupervised Domain Adaptation (UDA), which updates models to new situations without needing labeled data. They compare how much energy UDA uses versus training models from scratch, especially when labeling data is costly. The authors also suggest a way to decide when UDA is more energy-efficient than retraining, based on how many new domains appear. Their work helps understand the balance between energy use and labeling effort in adapting wireless network models.
6G wireless networksdistribution shiftsdata-driven modelsUnsupervised Domain Adaptationenergy consumptionmodel retraininglabeling costdomain adaptationmachine learning deploymentoptimization
Authors
Illyyne Saffar, Aurélie Boisbunon, Shruti Bothe
Abstract
The deployment of data-driven models in 6G wireless networks is increasingly challenged by frequent distribution shifts that degrade performance over time. Unsupervised Domain Adaptation (UDA) offers an alternative approach by adapting the trained model to a shifted domain without requiring labels. However, UDA pipelines are often more complex than single-task training due to additional modules and optimization procedures, raising a practical question: do the benefits of adaptation come at a higher energy cost, and how does this trade-off compare to retraining when labeling effort is also considered? In this work, we investigate the energy consumption of UDA and compare it to single task. We further propose a way to determine the minimum number of target domains for which UDA becomes more energy-efficient than retraining, taking into account the labeling cost. Our results aim to clarify when UDA should be preferred over classical train-from-scratch approaches from an energy and labeling-aware perspective.