JetBrains Research unites scientists working in challenging new disciplines

Application of machine translation methods to source code correction

In recent times, more and more machine learning techniques find their application to various code analysis tasks. Such tasks include: autocomplete, commit descriptions generation, code commenting, duplicate search etc.

Neural machine translation is currently a state-of-the-art method for automatic translation. We will discuss how to apply this method to source code to "translate" from an incorrect code to correct one.

Speaker: Artem Lobanov.

Presentation language: Russian.

Date and time: February 27th, 20:00-21:30.

Location: Times, room 204.

Videos from previous seminars are available at