"Pruning Neural Networks at Initialization: Why are We Missing the Mark?"

Jonathan Frankle et al. (2020)
a service of  Schloss Dagstuhl - Leibniz Center for Informatics