Mathematical models and artificial intelligence to understand natural phenomena: from the 'ALPS' project, funded by the European Research Council (ERC), a new approach for developing algorithms for the automated learning of mathematical models

The synergy between mathematical models and artificial intelligence will enable scientists to extract, in the form of equations, the laws governing natural phenomena that are still poorly understood. This is the goal of the ALPS project ‘AI-based Learning for Physical Simulation' funded by the European Research Council (ERC) through the ERC Starting Grant program. The project is coordinated by Alessandro Lucantonio, associate professor at The BioRobotics Institute, Sant’Anna School. "The synergy between mathematics as a natural language for physics and artificial intelligence algorithms is the main innovative feature of this approach that will enable increasingly accurate and understandable models."

Algorithms to understand reality 

The project aims to create new algorithms for automatically learning mathematical models for physics from experimental data. These new methods, based on a combination of Artificial Intelligence (AI) techniques, mathematical theories and advanced simulation tools, will enable scientists to develop models for understanding how reality works around us. In particular, mathematical theories provide a grammar and a set of concepts that will be combined by algorithms to build models of physical systems.
“Mathematics is the language of nature and scientists are able to unlock its secrets through the interpretation of mathematical models” explains Lucantonio.

Application Areas

The use of modern high-performance scientific computing tools is crucial to the success of the methods proposed in the project. In the long term, the results of the project will help researchers to address current challenges in the fields of robotics and medicine, such as the development of accurate models for the prediction of tumor growth.

"I am very excited about the Starting Grant obtained by Prof. Lucantonio," says Christian Cipriani, director of The BioRobotics Institute, "this funding will allow Alessandro to increase his international leadership in the field of computational modeling, including in the biorobotics. Then I am very proud for our Institute because we are winners of 4 active ERC projects. Good job to Alessandro!"