Description
Interesting Papers:
Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves
Metz et al. (2020)
M2SGD: Learning to Learn Important Weights
Kuo et al. (2020)
Meta-Learning in Neural Networks: A Survey
Hospedales et al., (2020)
Hyper-Meta Reinforcement Learning with Sparse Reward
Hua et al., (2020)
Meta-Curvature
Park et al., (2020)
Learning to Learn via Self-Critique
Antoniou et al., (2019)
La-MAML: Look-ahead Meta Learning for Continual Learning
Gupta et al. (2020)
Lectures:
https://www.youtube.com/watch?v=CRHKgOYXVe8
Useful libraries:
Learn2Learn - https://github.com/learnables/learn2learn
Higher - https://github.com/facebookresearch/higher
pytorch-meta - https://github.com/tristandeleu/pytorch-meta
cheers!