ML method · Self-supervised
Self-Supervised Learning
Training a model on unlabelled data by inventing a task it can grade itself on — predicting a hidden part of an image, or telling two views of the same object apart. It is powerful in astronomy, where raw data is abundant but expert labels are scarce.
Highlights
- ›Learning from abundant data without labels
Methods, data & targets
Used across astro-ML
Knowledge connections
- Associated withRepresentation Learning
- Associated withRepresentation Learning
- Associated withFoundation Models
- Associated withFoundation Models
Sources
The primary and reference sources this topic draws on.
- NASANational Aeronautics and Space Administration
Mission data, planetary science, space telescopes, and public-domain imagery.
Most NASA-produced imagery is in the public domain; individual items are checked for usage terms before publication.