Astronomy16 papers
Artificial Intelligence and Machine Learning in Planetary Science
A companion collection for the ACM 2026 talk by Valerio Carruba
VC
Valerio Carruba
All Reviews
AstronomyWorth Reading
intermediate
Machine learning classification of Kuiper belt populations
Smullen, Rachel A. & Volk, Kathryn (2020)
- Published
- Jul 6, 2020
- Journal
- Monthly Notices of the Royal Astronomical Society · Vol. 497 · No. 2
- DOI
- 10.1093/mnras/staa1935
At a GlanceAI
Machine learning classifies Kuiper belt objects into major dynamical groups with over 97% accuracy and far less effort.
SummaryAI
Classifying Kuiper belt objects usually requires lengthy orbital simulations and expert inspection, which will become harder as new surveys discover many more objects. This study shows that a Gradient Boosting Classifier trained on features from short numerical simulations can sort objects into four major dynamical populations with over 97% test accuracy. The authors also use the speed of the method to assess class membership across orbital-error clones, helping identify ambiguous cases and gaps in the training data. The result is a fast, scalable path for organizing the growing census of outer Solar System objects.
Method SnapshotAI
The study applies a Gradient Boosting machine-learning classifier to features extracted from short orbital simulations.
BackgroundAI
Basic background in Solar System dynamics, Kuiper belt populations, and introductory machine learning is helpful.
One of the first studies to apply machine learning to the dynamical classification of Kuiper Belt Objects, this work showed that gradient-boosting methods can provide fast and accurate classifications with minimal loss of reliability. It remains an important reference for researchers interested in scalable approaches to the analysis of the rapidly growing KBO population.
— VC