Textures and patterns are ubiquitous in astronomical data but challenging to quantify. The speaker will present a novel statistical tool, called the “scattering transform”. It borrows ideas from convolutional neural nets (CNNs), but retains advantages of traditional statistical estimators. As an example, he will show its application in weak lensing cosmology, where it outperforms classic statistics. It is a powerful new approach in astrophysics and beyond.
If time allows, the speaker will also briefly talk about another topic: he will present two new populations of white dwarfs revealed by Gaia data.
BIO
Sihao Cheng (程思浩) is a fourth-year graduate student at Johns Hopkins University, after graduating from Peking University. He works with prof. Brice Ménard on statistical analyses of survey data.
Host: Shude Mao
