Multi-Fidelity SINDy: Sparse Discovery of Nonlinear Dynamical Systems with Fidelity-Weighted Measurements
arXiv:2606.15690v1 Announce Type: new Abstract: Data from simulations and experiments are rarely noise-free and often exhibit heterogeneous levels of fidelity. Measurement uncertainty may vary across repeated observations, sensing devices, or even within a single experiment. This work addresses the problem of discovering nonlinear dynamical systems from such inhomogeneous data. We extend the Sparse Identification of Nonlinear Dynamical Systems (SINDy) framework to account for variable noise levels by combining Ensemble SINDy and Weak SINDy within a weighted regression formulation derived from
The demand for robust AI models that can learn from imperfect, real-world data is rapidly increasing as AI systems move from laboratories to deployed applications.
This development allows for more accurate and reliable discovery of physical laws and system dynamics from noisy, variable-fidelity data, which is crucial for building resilient AI for scientific discovery and control.
The ability of AI to interpret and learn from inconsistent or uncertain data sources is significantly enhanced, enabling more robust control systems and scientific modeling even with real-world complexities.
- · AI/ML researchers
- · Scientific computing sector
- · Robotics and automation
- · Complex systems modeling
- · Traditional data cleaning services (potentially less critical for some tasks)
- · Systems relying solely on high-fidelity, pristine datasets
Improved accuracy and reliability of AI models trained on heterogeneous, noisy data.
Accelerated scientific discovery and engineering innovation due to more robust system identification.
New classes of autonomous systems capable of operating reliably in highly uncertain and dynamic environments.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG