Domain Adaptation Under Wireless Network Constraints: When Does It Become Green?
Illyyne Saffar, Aurélie Boisbunon, Shruti Bothe
stat.ML, cs.AI, cs.LG
Submitted: 2026-06-22
License: http://creativecommons.org/licenses/by/4.0/
The gist: The deployment of data-driven models in 6G wireless networks is increasingly challenged by frequent distribution shifts that degrade performance over time.
Terminology
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.
Sources
- Carbontracker: Tracking and Predicting the Carbon Footprint of Training Deep Learning Models
- Through the telecom lens: Are all training samples important?
- Quantifying the Carbon Emissions of Machine Learning
- Carbon Emissions and Large Neural Network Training
- Few-Shot Domain Adaptation For End-to-End Communication
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Scalable Posterior Uncertainty for Flexible Density-Based Clustering
- A Transport Geometry of Benign Overfitting
- Efficient quadratic entropy with distance sketches