YerevaNN

research lab

YerevaNN /jɛɾɛvɑnˈɛn/ is a non-profit computer science and mathematics research lab based in Yerevan, Armenia.

Research

  • Time series prediction on medical data
  • Relation extraction from biological literature
  • Machine learning theory
    • T-Corex: a novel method for temporal covariance estimation using information theoretic apparatus [preprint] [code]
    • This is a collaboration with USC ISI.
  • Development of Armenian treebanks
    • Developing Eastern Armenian treebank according to the Universal Dependencies specifications [website] [code]
    • The latest version with 36K tokens is included in UD v2.4 [data]
    • An end-to-end syntax parser for Armenian [code] [demo]
    • This is a collaboration with Marat Yavrumyan and Anna Danielyan from Yerevan State University. Partially funded by ANSEF and ISTC.
  • Student projects
    • Morpheme-aware word vectors [code]
    • A joint POS tagger and lemmatizer [paper] [code]
    • Reproducing DIIN network for NLI [report] [code]
    • Reproducing R-NET network for QA [blogpost] [code]
Visit our blog and GitHub for more.

Team

Hrant Khachatrian / Github
Karen Hambardzumyan / Github
Tigran Galstyan / Github
Arshak Minasyan / Github
Ashot Matevosyan (intern) / Github
Past members: Hrayr Harutyunyan, Gor Arakelyan, Martin Mirakyan The easiest way to join YerevaNN is to start working on one of the projects listed on MLEVN Project Ideas Page. Our guide to deep learning might be useful (last updated in December, 2016)

The handwritten digits in the background are generated by deep convolutional generative adversarial networks [paper] [code]