Target Parameterization in Diffusion Models for Nonlinear Spatiotemporal System Identification

ArXi:2604.17566v1 Announce Type: cross Machine learning is becoming increasingly important for nonlinear system identification, including dynamical systems with spatially distributed outputs. However, classical identification and forecasting approaches become markedly less reliable in turbulent-flow regimes, where the dynamics are high-dimensional, strongly nonlinear, and highly sensitive to compounding rollout errors.