Neural Differential Equations combine neural networks with differential equations to model dynamic systems.
In this process, the rate of change of a state is modeled as a function of a neural network, which is particularly useful for continuous-time processes.
A well-known example is Neural Ordinary Differential Equations (Neural ODEs), where a neural network defines the derivative of a state vector and is numerically integrated.
These models enable the creation of flexible yet physically consistent predictions for time series, control problems, or physical simulations.