Interpretable Human-Label-Free Deep Learning for Real-Bogus Classification with Uncertainty Quantification

Raphaël Bonnet-Guerrini, Bruno Sanchez, Dominique Fouchez, Benjamin Racine, Maya Guy et al.|July 6, 2026arXiv

Key Takeaway

You can train accurate astronomical classifiers without expensive human labels by combining synthetic data injection with robust handling of noisy labels, and get reliable confidence scores through a hybrid uncertainty approach.

Summary

This paper develops a Real-Bogus classification system for astronomical transients that requires no human-labeled training data. It uses simulated transient injections combined with noisy survey data and a dual-network training approach to reliably distinguish real astronomical events from false detections, while also providing calibrated uncertainty estimates.

training evaluation safety

Key Terms

uncertainty-quantification synthetic-data label-noise co-teaching calibration