\emph{FoQuS}: A Forgetting-Quality Coreset Selection Framework for Automatic Modulation Recognition
This work addresses efficiency challenges for researchers and practitioners in signal processing and machine learning, though it is incremental as it builds on existing coreset selection methods.
The paper tackles the problem of high time and energy consumption in training deep learning models for Automatic Modulation Recognition by proposing FoQuS, a coreset selection framework that reduces training overhead while maintaining high recognition accuracy and cross-architecture generalization using only 1%-30% of the original data.
Deep learning-based Automatic Modulation Recognition (AMR) model has made significant progress with the support of large-scale labeled data. However, when developing new models or performing hyperparameter tuning, the time and energy consumption associated with repeated training using massive amounts of data are often unbearable. To address the above challenges, we propose \emph{FoQuS}, which approximates the effect of full training by selecting a coreset from the original dataset, thereby significantly reducing training overhead. Specifically, \emph{FoQuS} records the prediction trajectory of each sample during full-dataset training and constructs three importance metrics based on training dynamics. Experiments show that \emph{FoQuS} can maintain high recognition accuracy and good cross-architecture generalization on multiple AMR datasets using only 1\%-30\% of the original data.