Fetching the paper…

Sparse Networks from Scratch: Faster Training without Losing Performance · Around