Abstract
Machine-learning models for physical systems are currently evaluated primarily through errors between predicted and reference states and, increasingly, through tests of physical consistency. These metrics assess whether predictions are accurate and satisfy selected physical requirements, but provide limited insight into whether learned trajectories reproduce the underlying dynamics. Domain experts examine such relationships through physical representations that expose relevant processes, interactions, and responses, but these analyses are often separated from typical machine-learning evaluation. We introduce a practical framework for evaluating dynamical fidelity through predictive structure in physical representation spaces. Experts define the representations, while reference trajectories determine which relationships are predictive and retained as evaluation tests. We demonstrate the approach in atmospheric forecasting using ERA5 representations of planetary-wave activity and Northern Annular Mode evolution, and evaluate Pangu-Weather, GraphCast, and FengWu. The models exhibit distinct departures from reference predictive structure that are not reflected by conventional forecast errors. The framework thereby turns domain-expert representations into systematic tests of learned physical dynamics without prescribing the relationships in advance.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Lassen, O. B., Boger, J. P. D. S., Driscoll, S., Thomson, S. I., Schemm, S., Rodrigues, F., & Pereira, F. C. (2026). Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations. https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations
MLA 9
Lassen, Oskar Bohn, et al. "Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations." https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations.
Chicago (author–date)
Lassen, Oskar Bohn, Joao Paulo de Souza Boger, Simon Driscoll, Stephen I. Thomson, Sebastian Schemm, Filipe Rodrigues, and Francisco C. Pereira. 2026. "Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations." https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations.
Harvard
Lassen, O. B., Boger, J. P. D. S., Driscoll, S., Thomson, S. I., Schemm, S., Rodrigues, F. and Pereira, F. C. (2026) 'Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations', Available at: https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations.
Vancouver
Lassen OB, Boger JPDS, Driscoll S, Thomson SI, Schemm S, Rodrigues F, et al. Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations. https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations
IEEE
O. B. Lassen, J. P. D. S. Boger, S. Driscoll, S. I. Thomson, S. Schemm, F. Rodrigues, and F. C. Pereira, "Evaluating Dynamical Fidelity through Predictive Structure in Physical Representations," https://omanscience.com/en/articles/evaluating-dynamical-fidelity-through-predictive-structure-in-physical-representations.