Ongoing and planned weak lensing (WL) surveys are becoming deep enough to contain information on angular scales down to a few arcmin. To fully extract information from these small scales, we must capture non-Gaussian features in the cosmological WL signal while accurately accounting for baryonic effects. In this work, we account for baryonic physics via a baryonic correction model that modifies the matter distribution in dark matter-only $N$-body simulations, mimicking the effects of galaxy formation and feedback. We implement this model in a large suite of ray-tracing simulations, spanning a grid of cosmological models in $\Omega_\mathrm{m}-\sigma_8$ space. We then develop a convolutional neural network (CNN) architecture to learn and constrain cosmological and baryonic parameters simultaneously from the simulated WL convergence maps. We find that in a Hyper-Suprime Cam (HSC)-like survey, our CNN achieves a 1.7$\times$ tighter constraint in $\Omega_\mathrm{m}-\sigma_8$ space ($1\sigma$ area) than the power spectrum and 2.1$\times$ tighter than the peak counts, showing that the CNN can efficiently extract non-Gaussian cosmological information even while marginalizing over baryonic effects. When we combine our CNN with the power spectrum, the baryonic effects degrade the constraint in $\Omega_\mathrm{m}-\sigma_8$ space by a factor of 2.4, compared to the much worse degradation by a factor of 4.7 or 3.7 from either method alone.
Comment: 12 pages, 6 figures