The classifier judged this item not relevant to AI security, so it is left out of lists, feeds, the API and dataset releases. If that is wrong, report a correction on this page.
InfoResearchPeer-reviewed
Accurate and Robust Neural Architecture Search via a Flexible Supernet
- Published
- Record updated
Summary
ARNAS++ is a neural architecture search method that finds architectures balancing natural accuracy and adversarial robustness. It uses a supernet with flexible parameter budgets and width, controlled by a parameter budget controlling loss and a learnable filter number reduction ratio. The authors report outperforming competitors on six benchmark datasets under various adversarial attacks.