Exploring the universe with AI Kevin Schawinski Institute for - PowerPoint PPT Presentation

Exploring the universe with AI Kevin Schawinski Institute for Particle Physics and Astrophysics ETH Zurich ETH black hole group @kevinschawinski Grüp Bœgg Negar Politecnic da Zürig

how can machine learning/ artificial intelligence help us understand the universe?

“BIG DATA”

GalaxyGAN: de-noising and feature reconstruction PSFGAN: point source subtraction Generative models: data-driven exploration

generative adversarial network for overcoming limitations in astrophysical images Data Prep. Training of GAN Original Image Original Image (Original Image, Degraded Image) or (Recovered Image, Degraded Image) Discriminator Artificial Degrading Degraded Image Generator Recovered Image Schawinski+17

original degraded GAN recovered deconvolved PSF=2.5”, 10 σ

Training Architecture Original Discriminator Preprocessing Original + AGN Recovered Generator Dominik Stark PSFGAN, Stark+ submitted

PSFGAN, Stark+ submitted

Less sensitive to PSF changes Better at recovering features PSFGAN, Stark+ submitted

original data encoder latent space decoder reconstructed data z Dennis Turp

z 1 z 2 z’ = a × z 1 + b × z 2

z 1 z 2 z’ = a × z 1 + b × z 2

original data encoder latent space decoder reconstructed data z z SSFR age z reconstructed galaxies with reconstructed faces with original face SSFR changed in latent space age changed in latent space original galaxy

changing SSFR changing bulge-to-disk in latent space in latent space

machine learning can help us do better science by better understanding the data we have, and will get in the future go to space.ml to try out our projects!