MSL: Mining for Substructure Lenses

Discussion topics for individual codes
Post Reply
Ada Coda
ASCL Robot
Posts: 1702
Joined: Thu May 08, 2014 5:37 am

MSL: Mining for Substructure Lenses

Post by Ada Coda » Wed Sep 30, 2020 10:50 pm

MSL: Mining for Substructure Lenses

Abstract: MSL applies simulation-based inference techniques to the problem of substructure inference in galaxy-galaxy strong lenses. It leverages additional information extracted from the simulator, then trains neural networks to estimate likelihood ratios associated with population-level parameters characterizing dark matter substructure. The package including five high-level scripts which run the simulation and create samples, combing multiple simulation runs into a single file to use for training, then train the neural networks. After training, the estimated likelihood ratio is tested, and calibrated network predictions are made based on histograms of the network output.

Credit: Mishra-Sharma, Siddharth; Brehmer, Johann; Louppe, Gilles

Site: ... cture-lens

Bibcode: 2020ascl.soft09024M

Preferred citation method:

ID: ascl:2009.024

Post Reply