Tutorial on Machine Translation and Summarisation

Inigo Jauregi, Jacob Parnell and Massimo Piccardi

This tutorial addresses neural machine translation (NMT) and summarisation by deep learning approaches*. We will first provide a general introduction covering baseline models, training, inference and evaluation. We will then review some of the contemporary approaches, including reinforcement learning-based, transfer learning-based, unsupervised, and a few more. The rest of the tutorial will present code in PyTorch Lightning, a framework meant to remove the "boilerplate" from the development of research models. The speakers are Iñigo Jauregi Unanue, Jacob Parnell and Massimo Piccardi from UTS and RoZetta Technology.

*Yes, the name "neural summarisation" doesn't seem to really have cracked through. On the other hand, the "neural" in front of machine translation is being increasingly dropped, as if it's become a given.