Beyond Reptile: Meta-Learned Dot-Product Maximization between Gradients for Improved Single-Task Regularization.

Akhil Kedia, Sai Chetan Chinthakindi, Wonho Ryu

Browse the full EMNLP paper archive.