Speaker
Prof.
Xiao-Dong Li
(SYSU)
Description
We use the deep CNN to map the complex, non-linear cosmic large-scale structure to the underlying cosmological parameters $\Omega_m$, $\sigma_8$. In the ideal case, we can achieve a constraint 6 times better than the Planck CMB results, just using a $(256 \rm Mpc/h)^3$ box of dark matter density field with 2 Mpc/h resolution. Machine learning may play a very important role in the future astronomical data mining.
Primary author
Prof.
Xiao-Dong Li
(SYSU)
