AutoML.org

Freiburg-Hannover-Tübingen

NAS Best Practices

The rapid development of new NAS approaches makes it hard to compare these against each other. To ensure reliable and reproducible results, we also provide best practices for scientific research on NAS and our checklist for new NAS papers. In our CVPR 2021 NAS Workshop paper, we discuss some practical considerations that help improve the stability, efficiency, and overall performance of neural architecture search methods.

Reference