Neural Decipherment research named Netexplo/UNESCO award winner

Neural Decipherment

CSAIL research on “Translating lost languages using machine learning” has been announced as one of the 2021 Netexplo Award Winners in partnership with UNESCO. 

The Netexplo Observatory focuses on “groundbreaking digital innovations that can have a profound and lasting impact on the digital society of today and tomorrow.” 

MIT professor Regina Barzilay and MIT PhD student Jiaming Luo made a major development in the field: a new system that has been shown to be able to automatically decipher a lost language, without needing advanced knowledge of its relation to other languages. They also showed that their system can itself determine relationships between languages, and they used it to corroborate recent scholarship suggesting that the language of Iberian is not actually related to Basque.

“For undeciphered languages in real life, knowing what their related languages are is a crucial step towards decipherment. Our model can produce a ‘closeness’ metric that measures how much two vocabularies match,” says Luo. “More broadly, I believe that artificial intelligence (AI) can benefit from domain expertise in other fields. And vice versa, as fundamental scientific and human research can also benefit from AI innovations.”

Barzilay and Luo highlighted their work at the Netexplo Innovation Forum, and it was extensively covered by news media outlets such as the BBC, Fortune, CNET, VentureBeat.