OKC colloquia

Joint OKC-KVA-JSPS Colloquium: Machine Learning for Observational Cosmology

by Naoki Yoshida

FA32 (AlbaNova Main Building)


AlbaNova Main Building


Forthcoming wide-field sky surveys are expected to deliver a sheer volume of data exceeding an exabyte. Processing the large amount of multiplex astronomical data is technically challenging, and fully automated technologies based on machine learning and artificial intelligence are urgently needed. We present the results of our real data analysis with Subaru HSC survey. The applications include detection of transients, classification of supernovae, weak lensing analysis and cosmological parameter estimation, and emulator development for statistical analyses. Exciting prospects for future surveys using ground-based and space-borne telescopes will be given to conclude the  talk.