SEP 9, 2026 · PREPRINT
Physics-Informed Multi-Task Surrogate Model for the Martian Nightside Thermosphere
arXiv
This is a methods/proof-of-concept study demonstrating a physics-informed machine-learning approach to modeling Martian thermosphere data, with no clinical or human health outcome and no comparison to an established standard or competing method.
PreliminarySergey Nikiforov
Reported
Observation periodMY 32–38, 2014–2025
Species modeled4 (O, CO₂, N₂, Ar)
Primary evaluation metricsRMSE, MAE, R²