Search

We build a deep learning framework that connects the local formation process of dark matter haloes to the halo bias. We train a convolutional neural network (CNN) to predict the final mass and concentration of dark matter haloes from the initial conditions. The CNN is then used as a surrogate model to derive the response of the haloes' mass and concentration to long-wavelength perturbations in the initial conditions, and consequently the halo bias parameters following the 'response bias' definition. The CNN correctly predicts how the local properties of dark matter haloes respond to changes in the large-scale environment, despite no explicit knowledge of halo bias being provided during training. We show that the CNN recovers the known trends for the linear and second-order density bias parameters b 1 and b 2, as well as for the local primordial non-Gaussianity linear bias parameter b ϕ. The expected secondary assembly bias dependence on halo concentration is also recovered by the CNN: at fixed mass, halo concentration has only a mild impact on b 1, but a strong impact on b ϕ. Our framework opens a new window for discovering which physical aspects of the halo's Lagrangian patch determine assembly bias, which in turn can inform physical models of halo formation and bias. [ABSTRACT FROM AUTHOR]

We build a deep learning framework that connects the local formation process of dark matter haloes to the halo bias. We train a convolutional neural network (CNN) to predict the final mass and concentration of dark matter haloes from the initial conditions. The CNN is then used as a surrogate model to derive the response of the haloes' mass and concentration to long-wavelength perturbations in the initial conditions, and consequently the halo bias parameters following the 'response bias' definition. The CNN correctly predicts how the local properties of dark matter haloes respond to changes in the large-scale environment, despite no explicit knowledge of halo bias being provided during training. We show that the CNN recovers the known trends for the linear and second-order density bias parameters b 1 and b 2, as well as for the local primordial non-Gaussianity linear bias parameter b ϕ. The expected secondary assembly bias dependence on halo concentration is also recovered by the CNN: at fixed mass, halo concentration has only a mild impact on b 1, but a strong impact on b ϕ. Our framework opens a new window for discovering which physical aspects of the halo's Lagrangian patch determine assembly bias, which in turn can inform physical models of halo formation and bias. [ABSTRACT FROM AUTHOR]

The integrated shear 3-point correlation function ζ± is a higher-order statistic of the cosmic shear field that describes the modulation of the 2-point correlation function ξ± by long-wavelength features in the field. Here, we introduce a new theoretical model to calculate ζ± that is accurate on small angular scales, and that allows to take baryonic feedback effects into account. Our model builds on the realization that the small-scale ζ± is dominated by the non-linear matter bispectrum in the squeezed limit, which can be evaluated accurately using the non-linear matter power spectrum and its first-order response functions to density and tidal field perturbations. We demonstrate the accuracy of our model by showing that it reproduces the small-scale ζ± measured in simulated cosmic shear maps. The impact of baryonic feedback enters effectively only through the corresponding impact on the non-linear matter power spectrum, thereby permitting to account for these astrophysical effects on ζ± similarly to how they are currently accounted for on ξ±. Using a simple idealized Fisher matrix forecast for a DES-like survey we find that, compared to ξ±, a combined |$\xi _{\pm }\ \&\ \zeta _{\pm }$| analysis can lead to improvements of order |$20\!-\!40{{\ \rm per\ cent}}$| on the constraints of cosmological parameters such as σ8 or the dark energy equation of state parameter w 0. We find similar levels of improvement on the constraints of the baryonic feedback parameters, which strengthens the prospects for cosmic shear data to obtain tight constraints not only on cosmology but also on astrophysical feedback models. These encouraging results motivate future works on the integrated shear 3-point correlation function towards applications to real survey data. [ABSTRACT FROM AUTHOR]