SoftServe: A Scalable Quasi-Newton Method for Deep Learning
SoftServe scales positive-definite quasi-Newton updates to large networks and beats Adam, Muon, and SOAP on ill-conditioned tasks.
1 stories, newest first · one card per story
SoftServe scales positive-definite quasi-Newton updates to large networks and beats Adam, Muon, and SOAP on ill-conditioned tasks.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views