Yandex launches machine intelligence and research division - RUSSOFT

Supported by:

Yandex launches machine intelligence and research division

This division is defined as a “centralized, cross-functional unit to accelerate innovation and unify [Yandex’s] core machine learning technologies”

Feb 15, 2017
Earlier this week Yandex, the leading Internet search company in Russian-language countries, announced the creation of its new "Machine Intelligence and Research (MIR) Division."

The MIR division will also transfer research from various Yandex research teams into the company’s products and services.

Research will focus on AI-centered technologies including:
  • Machine learning, using such existing in-house technologies as MatrixNet – which is now used in nearly every Yandex product and service – and deep neural network framework DaNet;
  • Computer vision, a field in which Yandex operates a market-leading image recognition technology, using machine learning to detect similar images in visual search results and perform a number of high-end vision tasks;
  • Speech, with the voice recognition technology SpeechKit, which powers voice commands for Yandex’s search and traffic information applications, and has been made available to the company’s corporate clients;
  • Translation, with more than 90 languages in production, making Yandex "one of very few companies in the world that has access to enough data to meet today’s high machine translation standards."
Misha Bilenko has been appointed to head the new division. He previously led Microsoft’s Machine Learning Algorithms team in the Cloud and Enterprise division, following a career in the Machine Learning Group for Microsoft Research.