Deep neural networks using temperature-based spectral representations recover planetary Doppler signals with amplitudes of at least 25 cm/s from HARPS-N solar spectra under cross-validation.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Generative spectral decomposition on NEID solar data recovers 238 of 500 injected planets (including 13 with K<0.3 m/s) versus 9 for traditional CCF detrending in calibrated injection-recovery tests.
Solar image analysis shows typical astrometric jitter of 0.342 μas pc from activity, below the ~3 μas Earth-planet signal at 1 pc, so stellar contamination does not prevent Earth-like exoplanet astrometry around Sun-like stars.
citing papers explorer
-
Modeling Doppler Shifts in Radial-Velocity Data with Deep Learning toward Earth-mass Exoplanet Detection
Deep neural networks using temperature-based spectral representations recover planetary Doppler signals with amplitudes of at least 25 cm/s from HARPS-N solar spectra under cross-validation.