Global Convergence and Induced Kernels of Gradient-Based Meta-Learning with Neural Nets

Open in new window