# End-to-End Training for Back-Translation with Categorical Reparameterization Trick

arXiv: https://arxiv.org/abs/2202.08465 · HTML: https://mlsanity.com/paper/2202.08465/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 27 of its 43 references that are in the map.
- First posted on arXiv 17 Feb 2022.

## Builds on (most cited first, 27 of 27)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [A Call for Clarity in Reporting BLEU Scores](https://mlsanity.com/paper/1804.08771/) (2018)
- [fairseq: A Fast, Extensible Toolkit for Sequence Modeling](https://mlsanity.com/paper/1904.01038/) (2019)
- [XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks](https://mlsanity.com/paper/1603.05279/) (2016)
- [Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks](https://mlsanity.com/paper/1506.03099/) (2015)
- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [Disentangling by Factorising](https://mlsanity.com/paper/1802.05983/) (2018)
- [Scaling Neural Machine Translation](https://mlsanity.com/paper/1806.00187/) (2018)
- [On Using Monolingual Corpora in Neural Machine Translation](https://mlsanity.com/paper/1503.03535/) (2015)
- [Dual Learning for Machine Translation](https://mlsanity.com/paper/1611.00179/) (2016)
- [Tagged Back-Translation](https://mlsanity.com/paper/1906.06442/) (2019)
- [Unsupervised Neural Machine Translation with Generative Language Models Only](https://mlsanity.com/paper/2110.05448/) (2021)
- [Joint Training for Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1803.00353/) (2018)
- [Back-Translation Sampling by Targeting Difficult Words in Neural Machine Translation](https://mlsanity.com/paper/1808.09006/) (2018)
- [Revisiting Iterative Back-Translation from the Perspective of Compositional Generalization](https://mlsanity.com/paper/2012.04276/) (2020)
- [Generalizing Back-Translation in Neural Machine Translation](https://mlsanity.com/paper/1906.07286/) (2019)
- [Meta Back-translation](https://mlsanity.com/paper/2102.07847/) (2021)
- [Techniques for Learning Binary Stochastic Feedforward Neural Networks](https://mlsanity.com/paper/1406.2989/) (2014)
- [Explaining and Generalizing Back-Translation through Wake-Sleep](https://mlsanity.com/paper/1806.04402/) (2018)
- [Dual Reconstruction: a Unifying Objective for Semi-Supervised Neural Machine Translation](https://mlsanity.com/paper/2010.03412/) (2020)
- [Deep Neural Networks and End-to-End Learning for Audio Compression](https://mlsanity.com/paper/2105.11681/) (2021)

## Abstract (arXiv)

Back-translation (BT) is an effective semi-supervised learning framework in neural machine translation (NMT). A pre-trained NMT model translates monolingual sentences and makes synthetic bilingual sentence pairs for the training of the other NMT model, and vice versa. Understanding the two NMT models as inference and generation models, respectively, the training method of variational auto-encoder (VAE) was applied in previous works, which is a mainstream framework of generative models. However, the discrete property of translated sentences prevents gradient information from flowing between the two NMT models. In this paper, we propose the categorical reparameterization trick (CRT) that makes NMT models generate differentiable sentences so that the VAE's training framework can work in an end-to-end fashion. Our BT experiment conducted on a WMT benchmark dataset demonstrates the superiority of our proposed CRT compared to the Gumbel-softmax trick, which is a popular reparameterization method for categorical variable. Moreover, our experiments conducted on multiple WMT benchmark datasets demonstrate that our proposed end-to-end training framework is effective in terms of BLEU scores not only compared to its counterpart baseline which is not trained in an end-to-end fashion, but also compared to other previous BT works. The code is available at the web.