Researchers from the Institute of Cosmos Sciences of the University of Barcelona (ICCUB) have developed a new framework based on machine learning that significantly improves the resolution of complex differential equations, especially in cases where traditional methods present difficulties. The study, led by experts Pedro Tarancón-Álvarez and Pablo Tejerina-Pérez, has been published in the journal Communications Physics (Nature publishing group).
Differential equations are fundamental tools in physics: they are used to describe phenomena ranging from fluid dynamics to general relativity. But when these equations become stiff (i.e. they involve very different scales or highly sensitive parameters), they become extremely difficult to solve. This is especially relevant in inverse problems, where scientists try to deduce unknown physical laws from observed data.
