Alternating Synthetic and Real Gradients for Neural Language Modeling

02/27/2019
by   Fangxin Shang, et al.
2

Training recurrent neural networks (RNNs) with backpropagation through time (BPTT) has known drawbacks such as being difficult to capture longterm dependencies in sequences. Successful alternatives to BPTT have not yet been discovered. Recently, BP with synthetic gradients by a decoupled neural interface module has been proposed to replace BPTT for training RNNs. On the other hand, it has been shown that the representations learned with synthetic and real gradients are different though they are functionally identical. In this project, we explore ways of combining synthetic and real gradients with application to neural language modeling tasks. Empirically, we demonstrate the effectiveness of alternating training with synthetic and real gradients after periodic warm restarts on language modeling tasks.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/20/2017

Neural Networks Compression for Language Modeling

In this paper, we consider several compression techniques for the langua...
research
05/17/2019

Adaptively Truncating Backpropagation Through Time to Control Gradient Bias

Truncated backpropagation through time (TBPTT) is a popular method for l...
research
02/25/2016

Recurrent Neural Network Grammars

We introduce recurrent neural network grammars, probabilistic models of ...
research
11/23/2016

Scalable Bayesian Learning of Recurrent Neural Networks for Language Modeling

Recurrent neural networks (RNNs) have shown promising performance for la...
research
11/26/2015

Regularizing RNNs by Stabilizing Activations

We stabilize the activations of Recurrent Neural Networks (RNNs) by pena...
research
01/19/2021

Implicit Bias of Linear RNNs

Contemporary wisdom based on empirical studies suggests that standard re...
research
10/05/2020

Understanding the Mechanics of SPIGOT: Surrogate Gradients for Latent Structure Learning

Latent structure models are a powerful tool for modeling language data: ...

Please sign up or login with your details

Forgot password? Click here to reset