Skip to content (access key 's')
Logo of Technion
Logo of CS Department
Events

The Taub Faculty of Computer Science Events and Talks

Koopman-Based Stochastic Differential Equations for Time Series Forecasting
event speaker icon
Firas Yazbak (M.Sc. Thesis Seminar)
event date icon
Thursday, 16.04.2026, 13:00
event location icon
Taub 9 & Zoom
event speaker icon
Advisor: Prof. Assaf Schuster

Koopman-based methods for time-series forecasting model nonlinear dynamics as linear evolution in a latent space, but deterministic formulations often mix noise with the underlying dynamics, limiting long-term accuracy. This talk introduces KoopSDE, which extends Koopman models with a latent stochastic differential equation.

By linking the drift to the Koopman generator and learning a diffusion term, the method separates dynamics from noise and improves robustness in long-horizon forecasting.