Cameron Wolfe

Latest

• How Much Pre-training Is Enough to Discover a Good Subnetwork?
• Momentum-inspired low-rank coordinate descent for diagonally constrained SDPs