The Neural Noisy Channel
This work addresses the problem of improving sequence transduction models for NLP tasks by leveraging unpaired data, though it is incremental as it builds on existing noisy channel concepts with neural parameterization.
The paper tackles sequence-to-sequence transduction by formulating it as a noisy channel decoding problem using recurrent neural networks, and shows that this approach outperforms direct models on tasks like summarization and machine translation, with significant benefits from unpaired output data.
We formulate sequence to sequence transduction as a noisy channel decoding problem and use recurrent neural networks to parameterise the source and channel models. Unlike direct models which can suffer from explaining-away effects during training, noisy channel models must produce outputs that explain their inputs, and their component models can be trained with not only paired training samples but also unpaired samples from the marginal output distribution. Using a latent variable to control how much of the conditioning sequence the channel model needs to read in order to generate a subsequent symbol, we obtain a tractable and effective beam search decoder. Experimental results on abstractive sentence summarisation, morphological inflection, and machine translation show that noisy channel models outperform direct models, and that they significantly benefit from increased amounts of unpaired output data that direct models cannot easily use.