Obtaining ordinary differential equations (ODEs) models with symbolic polynomial neural ODEs

What is it about?

Ordinary differential equations (ODEs) are used to describe many systems in the social and natural sciences, such as physics, meteorology, astronomy, chemistry, biology, ecology, and economics. The process of obtaining ODE equations to describe the system often takes years of model development and refinement from observed data. We present a new technique to obtain an ODE model, which we call the symbolic polynomial neural ODEs.

Featured Image

Photo by Antoine Dautry on Unsplash

Photo by Antoine Dautry on Unsplash

Why is it important?

Obtaining ODE models is an important part of understanding how social and natural science systems work. Once a model is obtained, it is used to understand the processes underlying the system's dynamics. Using the knowledge obtained by the model, one can then try to perturb the system to get a desired outcome. For example, we can try to cure a disease if we understand the dynamics that cause the disease. We could also try to change the economic system in a favorable way if we have a model for the economy.