Think Again Networks and the Delta Loss
2019-04-26Code Available0· sign in to hype
Alexandre Salle, Marcelo Prates
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/jb33k/awd-lstm-lm-ThinkNetpytorch★ 0
Abstract
This short paper introduces an abstraction called Think Again Networks (ThinkNet) which can be applied to any state-dependent function (such as a recurrent neural network).