Thursday 23 January 2020, 10:00, LSV library: two talks, François Charton, Guillaume Lample (facebook).
Neural networks have a reputation for being better at solving statistical or approximate problems than at performing calculations or working with symbolic data. In this paper, we show that they can be surprisingly good at more elaborated tasks in mathematics, such as symbolic integration and solving differential equations. We propose a syntax for representing mathematical problems, and methods for generating large datasets that can be used to train sequence-to-sequence models. We achieve results that outperform commercial Computer Algebra Systems such as Matlab or Mathematica.
Machine translation (MT) has achieved impressive results recently, thanks to recent advances in deep learning and the availability of large-scale parallel corpora. Yet, their effectiveness strongly relies on the availability of large amounts of parallel sentences, which hinders their applicability to the majority of language pairs. Previous studies have shown that monolingual data -- widely available in most languages -- can be used to improve the performance of MT systems. However, these were used to augment, rather than replace, parallel corpora. In this talk, we will present our recent research on Unsupervised Machine Translation, where we show that it is possible to train MT systems in a fully unsupervised setting, without the need of any cross-lingual dictionary or parallel resources whatsoever, but only with access to large monolingual corpora in each language.