Machine Learning Validates Transient Astronomical Phenomena in Historical Images

This paper presents evidence supporting the existence of previously unrecognized transient astronomical phenomena in historical observatory images. The study addresses concerns that transients identified via existing automated pipelines are simply plate defects by using machine learning (ML) to enhance transient identification accuracy and validate the phenomenon. A machine learning model was trained against 250 transient image pairs classified as real versus plate defect by expert visual review. The model demonstrated good discrimination (out-of-fold AUC=0.81; sensitivity=0.71, specificity=0.71). After deployment in a dataset of 107,875 previously-identified transients, the model assigned each a probability of being real. After controlling for ML-identified artifacts, transient counts were significantly elevated for dates within a nuclear window (p=0.024). Transients with the highest probability of being real were more likely to occur within a nuclear window (p<0.0001). The shadow deficit was significant (p<0.0001) and largest in the highest probability transients relative to lower probability transients (p=0.003). The results strongly support the existence of an unrecognized population of transient objects in historical astronomical plates, warranting further study. The use of machine learning to analyze historical astronomical data represents a novel approach to astronomical discovery, potentially revealing new insights into the universe.

_Context: This intelligence report was compiled by the DailyOrbitalWire Strategy Engine. Verified for Art. 50 Compliance._