Image Details
Caption: Figure 6.
A comparison of the performance of BNNs using metrics averaged over the redshift range 0.3 < z < 1.5 (see Section 3.3.1). The gray shaded area corresponds to the LSST science goals. The models, with the exception of BNN-2, meet the same LSST bias and outlier requirements. Regardless of the training approach, these models fail to perform comparably on both ground truths.
© 2026. The Author(s). Published by the American Astronomical Society.