Page History

...

• Download and preprocess large hydrodynamic simulation data (i.e. TNG300).
• Implement and train a GNN to estimate dark matter halo masses from baryonic properties (starter code is available).
• Project 1
  • Augment GNN using added information from galaxy morphology classifications.
  • Augment GNN using convolutional neural network (CNN)-extracted morphological features from synthetic galaxy images.
• Project 2
  • Obtain galaxy catalog data for clusters (e.g. JWST galaxy catalogs for SMACS 0723 or Abell 2744).
  • Compare GNN predictions against lensing measurements.
• Project 3
  • Extract analytical equation for M_halo as a function of M_star and other environmental parameters by using symbolic regression.
  • Compare against other standard tools such as halo occupation distribution modeling.
• Write and publish a short paper detailing findings.

...