Bridging simulations and observations: New insights into galaxy formation simulations via out-of-distribution detection and Bayesian model comparison - Evaluating galaxy formation simulations under limited computing budgets and sparse dataset sizes

Open Access

Fig. 4

Average training and validation loss of each epoch during k-sparse VAE training on the SDSS training set and validating on the SDSS early stopping validation set. The vertical red line shows the lowest validation loss checkpoint and the vertical blue line indicates the epoch where our VAE model is saved. This early stopping mechanism avoids overfitting.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.