acm-header
Sign In

Communications of the ACM

ACM Careers

Machine Learning System Automatically Translates Long-Lost Languages


View as: Print Mobile App Share:
Linear B tablet

The researchers showed how machine translation can decipher languages that have been lost entirely.

Credit: N. Samartzidis

The study of linguistics has been revolutionized by the availability of huge annotated databases, and techniques for getting machines to learn from them. Consequently, machine translation from one language to another has become routine. These methods have provided an entirely new way to think about language.

Enter Jiaming Luo and Regina Barzilay from MIT and Yuan Cao from Google's AI lab in Mountain View, California. This team Researchers fromMIT and Google's AI lab have developed a machine-learning system capable of deciphering lost languages, and they've demonstrated it by having it decipher the ancient script Linear B—the first time this has been done automatically.

The key insight enabling machine translation is that words in different languages occupy the same points in their respective parameter spaces. That makes it possible to map an entire language onto another language with a one-to-one correspondence.

From Technology Review
View Full Article


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account