Training neural networks faster with minimal tuning using pre-computed lists of hyperparameters for NAdamW

Add code
Mar 06, 2025

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: