Russian students trained a neural network to forecast the movement of Arctic sea ice
The developers were offered internships at Skoltech's Centre for Applied AI, where they will be able to continue working on the project.6 September 2023
The neural network has been trained to identify the dynamics of Arctic ice movement, which will contribute to enhancing the safety of northern navigation. The system was created by Russian student winners of the Smiles 2023 hackathon, a summer school event focused on artificial intelligence (AI) development, which took place in the Altai Territory. Skoltech's press service reported this information to TASS.
According to the developers themselves, the project is also designed for sailors who need a precise understanding of ice movement patterns. 'Typically, they rely on satellite images, but these are not taken daily, which prevents the use of traditional machine learning methods,' quoted the press service from the developers' statement.
The peculiarity of the solution is that the system will identify patterns in the interaction of nearby ice patches and weather influences, thereby predicting ice movement dynamics. It is suggested that the model could be used in situations where there is a lack of actual data. The developers were offered internships at Skoltech's Centre for Applied AI, where they will be able to continue working on the project.
The project was one of the winners of the hackathon held as part of a summer school, which ran for 12 days in line with the goals of the federal project 'Artificial Intelligence' under the national project 'Digital Economy.'