Use unlearn methods to unlearn the trained sparse models
Model Sparsity Can Simplify Machine Unlearning: [https://github.com/OPTML-Group/Unlearn-Sparse]
SCRUB and SFRon: [https://github.com/K1nght/Unified-Unlearning-w-Remain-Geometry]
Unstructure: [https://github.com/rahulvigneswaran/Lottery-Ticket-Hypothesis-in-Pytorch]
Structure: [https://github.com/he-y/soft-filter-pruning]