Identifying and Controlling Important Neurons in Neural Machine Translation
This work addresses the interpretability and controllability of neural machine translation for researchers and practitioners, but it is incremental as it builds on existing knowledge about linguistic representations in NMT.
The authors tackled the problem of understanding and controlling neural machine translation models by developing unsupervised methods to identify important neurons, showing that translation quality depends on these neurons and that modifying their activations allows predictable control over translations.
Neural machine translation (NMT) models learn representations containing substantial linguistic information. However, it is not clear if such information is fully distributed or if some of it can be attributed to individual neurons. We develop unsupervised methods for discovering important neurons in NMT models. Our methods rely on the intuition that different models learn similar properties, and do not require any costly external supervision. We show experimentally that translation quality depends on the discovered neurons, and find that many of them capture common linguistic phenomena. Finally, we show how to control NMT translations in predictable ways, by modifying activations of individual neurons.