AI - Data - Apps - Cybernetics

Machine Learning As A Transformative Tool for (Exo-)Planetary Science

By Keith Cowing

Status Report

astro-ph.EP

April 13, 2026

Filed under AI, astro-ph.EP, autoencoder, Bayesian Analysis, data, datasets, Deep Neural Network, Machine learning, modelling, NCCR PlanetS

Machine Learning As A Transformative Tool for (Exo-)Planetary Science — Exoplanet detection limits from the neural network framework in the case of HARPS-N solar spectra. Left: Detection limit map in the period-amplitude domain. The red dots indicate the successful detection by the trained NN with a false alarm probability (FAP) > 0.1%. Middle: Amplitude comparison between the injected and recovered signals in the period-amplitude domain. The amplitude difference for most of the recovered signals are < 20%. The large difference at low amplitude (0.1 m/s) is likely due to noise in the data. Right: Phase comparison between the injected and recovered signals in the period-amplitude domain. The phase difference for most of the recovered planets is < 0.04. The large differences at long periods is likely due to a poorer sampling of the phase for long-period planetary signals.--astro-ph.EP

The exploration of planetary bodies in our Solar system and beyond relies on the processing and interpretation of large, spatio-temporally inconsistent, and heterogeneous datasets.

Recent advances in machine learning (ML) provide unprecedented opportunities to address many fundamental challenges posed by these heterogeneous and hyper-dimensional datasets. This review chapter highlights innovative ML methodologies that were developed and used by NCCR PlanetS members to address three overarching challenges in (exo)planetary science.

The first challenge is sequence modelling, which encompasses the intricate analysis of one-dimensional data such as time series of radial velocities and light curves, among other examples.

Secondly, there is pattern recognition that involves studying correlations, leveraging convolutional neural networks for feature extraction, mapping and cross correlation among other examples., anomaly detection through variational autoencoders, and unsupervised clustering of mass spectrometric data.

Lastly, there are generative models and emulation-based Bayesian analysis, which encompass the development of predictive models for planetary interior structure, employing Deep Neural Networks to understand planet formation mechanisms.

These innovative ML methodologies herald a paradigm shift in the processing of data and numerical models that represent inherent challenges in planetary and exoplanetary science, paving the way for revolutionary discoveries and ideas in this field.

J. Davoult, V. T. Bickel, C. Haslebacher, Y. Alibert, D. Angerhausen, C. Cantero, J. A. Egger, R. Eltschinger, Y. Eyholzer, E. O. Garvin, S. Gruchola, A. Leleu, S. Marques, Y. Zhao

Comments: Chapter accepted for publication in the NCCR PlanetS Legacy Book: Benz, W. et al. (Eds), The National Center for Competence in Research, PlanetS: A Swiss-wide network expanding planetary sciences. Springer (2026)
Subjects: Earth and Planetary Astrophysics (astro-ph.EP)
Cite as: arXiv:2604.09152 [astro-ph.EP] (or arXiv:2604.09152v1 [astro-ph.EP] for this version)
https://doi.org/10.48550/arXiv.2604.09152
Focus to learn more
Submission history
From: Jeanne Davoult PhD
[v1] Fri, 10 Apr 2026 09:39:14 UTC (12,913 KB)
https://arxiv.org/abs/2604.09152

Astrobiology, AI,

Keith Cowing

Biologist, Explorers Club Fellow, ex-NASA Space Biologist and Payload integrator, Editor of NASAWatch.com and Astrobiology.com, Lapsed climber, Explorer, Synaesthete, Former Challenger Center board member 🖖🏻

Follow on Twitter