Learning to learn by gradient descent by gradient descent

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Editable Neural Networks (2020, methodology): “(2018)), learnable optimization (Andrychowicz et al. (2016)) and reinforcement learning (Houthooft et al. (2018)). Indeed, Editable Neural Networks also belong to the meta-learning paradigm, as they basically ”learn to allow effective patching”. While neural network correction has significant practi”
Bayesian Recurrent Neural Networks (2017, background): “Lastly, learning to optimise (or learning to learn) (Li & Ma-lik, 2016; Andrychowicz et al., 2016) is related in that a learning rate is learned so that it produces better updates than those provided by e.g. AdaGrad (Duchi et al., 2011) or Adam (Kingma & Ba, 2014).”
Recurrent Inference Machines for Solving Inverse Problems (2017, methodology): “[15] that aims to train RNNs as optimizers for non-convex optimization problems.”
RNA-protein binding motifs mining with a new hybrid deep learning based cross-domain knowledge integration approach (2016, background): “exploit the structures in network architecture of interest [44].”
TRADI: Tracking deep neural network weight distributions (2019, background): “As proposed in the main article, and similarly with [1, 6], we assume that weights during back-propagation and forward pass are independent.”