A team of students in the UK have created a way to turn handwritten equations into computer code, which could help maths experts solve their most complex problems.
Dominik Henjes, Paul Popa and Aleksandr Jermakov came up with their algorithm during a hackathon that challenged people to use Artificial Intelligence (AI) and Machine Learning (ML) to improve lives.
Their system recognises equations and coverts them into code, even if they were written by hand on paper. It could be used by experts to develop their own programmes to solve difficult maths problems and eliminate mistakes when manually typing equations into a computer.
The University of Birmingham students won the hackathon, which was organised by digital services and platforms provider, Kainos, and held in the city earlier this month.
“Dominik, Paul and Aleksandr were worthy winners,” said Tom Gray, Chief Technology Officer for Kainos. “All the 20 undergraduates who took part showed a genuine passion for AI, which was inspirational to see.”
All the entries ran on Microsoft’s cloud platform, Azure, while Naveed Hussain, a Big Data and AI architect at the company, was one of the judges. Other solutions included using AI to detect bias in articles, accurately predict flight times and delays, estimate the amount of rubbish produced in the UK and predict streets where crime will occur.
Jermakov hailed the hackathon as “the perfect stepping stone from academic knowledge towards real-life applications of Artificial Intelligence.”