SOURCE-LINKED INTELLIGENCE

Sparsity-Adaptive Sharpness-Aware Minimization

arXiv · AI, language, vision and robotics · article · Sep 13, 2026 · UTC

Deploying deep neural networks in real-world settings requires models that are both compact and robust to common corruptions. However, at deployment-relevant high sparsity, standard pruning pipelines often degrade corruption robustness, and existing sharpness-aware training/pruning approaches provide limited robustness gains. We address this issue by introducing Sparsity-Adaptive Sharpness-Aware Minimization (SA-SAM), which derives a sparsity-dependent SAM/ASAM perturbation radius by keeping the mean absolute perturbation (an $\ell_1$-based proxy) approximately invariant as sparsity increases.

Read original source ↗ Open in workspace

recordType

paper

region

Global