A team of MSc and PhD students from Skoltech’s Advanced Data Analytics in Science and Engineering (ADASE) group placed third in the Data Science Game, a prestigious international competition held each year in Paris.
Some 340 teams from more than 250 universities in 40 different countries battled for victory in the game, and in the pre-final, Russian teams swept the top three spots. Moscow State University took first place, followed by the Higher School of Economics and Skoltech. In the final, Moscow State University scored first, followed by the University Pierre et Marie Curie (France) and Skoltech.
Each team created an algorithm designed to predict demand for automobile component parts based on various types of information, including supplier partners, product characteristics and order history.
As explained by ADASE team member Nikita Kluchnikov, this challenge raises a number of critical issues for industry.
“The balancing of supply and demand is the main problem in modern supply chains; precise demand forecasts for several months ahead can give a strong competitive advantage to a company due to more efficient inventory management and a predictable production schedule,” he said.
In addition to Kluchnikov, the ADASE team members include Ivan Nazarov, Vladislav Ishimtsev and Denis Volhonsky.
ADASE, run by Skoltech Professor Evgeny Burnaev, specializes in the development of machine learning methods for industrial applications.
Burnaev voiced enthusiasm about his team’s results: “This competition provided to us with a particularly valuable opportunity to test the algorithms we have developed and to compare their accuracy with those of the algorithms, developed by other participants.”
“I am proud of our team’s results, especially considering the fact that among our rivals were experienced machine learning competitors,” he added.
The Data Science Game has been held annually in Paris for the past three years.
Feature photo of Paris skyline: Moyan Brenn // Wikimedia Commons.