ASCL Code Record
[ascl:2401.010]
SYSNet: Neural Network modeling of imaging systematics in galaxy surveys
The Feed Forward Neural Network SYSNet models the relationship between the imaging maps, such as stellar density and the observed galaxy density field, in order to mitigate the systematic effects and to make a robust galaxy clustering measurements. The cost function is Mean Squared Error and a L2 regularization term, and the optimization algorithm is Adaptive Moment (ADAM).
- Code site:
- https://github.com/mehdirezaie/SYSNet
- Described in:
- https://ui.adsabs.harvard.edu/abs/2020MNRAS.495.1613R
- Bibcode:
- 2024ascl.soft01010R
Preferred citation method:
Views: 222
