Continuous-time models

Neural stochastic differential equation

| Artificial Intelligence |

Neural stochastic differential equations (Neural SDEs) combine stochastic differential equations with neural networks to model complex dynamical systems with inherent randomness. They extend classical neural ordinary differential equations by incorporating stochasticity, enabling richer representations of time series and continuous-time processes.

Read More →

Neural controlled differential equation

| Artificial Intelligence |

Neural controlled differential equations (Neural CDEs) are a class of machine learning models that generalize neural ordinary differential equations by incorporating control signals as inputs. They offer a continuous-time framework for modeling sequential data and have applications in time series analysis, physics-informed learning, and stochastic processes.

Read More →