Penalty and Augmented Lagrangian Methods for Layer-parallel Training of Residual Networks

Open in new window