A pruning-aware adaptive gradient method is proposed which classifies the variables in two sets before updating them using different strategies. This technique extends the ``relevant/irrelevant" approach of Ding (2019) and Zimmer et al. (2022) and allows a posteriori sparsification of the solution of model parameter fitting problems. The new method is proved to be convergent with a global rate of decrease of the averaged gradient's norm of the form \calO(log(k)/k+1). Numerical experiments on several applications show that it is competitive.
@article{arxiv.2502.08308,
title = {prunAdag: an adaptive pruning-aware gradient method},
author = {Margherita Porcelli and Giovanni Seraghiti and Philippe L. Toint},
journal= {arXiv preprint arXiv:2502.08308},
year = {2025}
}