Constraining primordial non-Gaussianity using Neural Networks

We present a novel approach to estimate the value of primordial non-Gaussianity ($f_{\mathrm{N}\mathrm{L}}$) parameter directly from the Cosmic Microwave Background (CMB) maps using a convolutional neural network (CNN). While traditional methods rely on complex statistical techniques, this study proposes a simpler approach that employs a neural network to estimate $f_{\mathrm{N}\mathrm{L}}$. The neural network model is trained on simulated CMB maps with known $f_{\mathrm{N}\mathrm{L}}$ in range of $[-50,50]$, and its performance is evaluated using various metrics. The results indicate that the proposed approach can accurately estimate $f_{\mathrm{N}\mathrm{L}}$ values from CMB maps with a significant reduction in complexity compared to traditional methods. With $500$ validation data, the $f_{\mathrm{N}\mathrm{L}}^{\mathrm{o}\mathrm{u}\mathrm{t}\mathrm{p}\mathrm{u}\mathrm{t}}$ against $f_{\mathrm{N}\mathrm{L}}^{\mathrm{i}\mathrm{n}\mathrm{p}\mathrm{u}\mathrm{t}}$ graph can be fitted as $y=ax+b$, where $a=0.980_{-0.102}^{+0.098}$ and $b=0.277_{-0.101}^{+0.098}$, indicating the unbiasedness of the primordial non-Gaussianity estimation. The results indicate that the CNN technique can be widely applied to other cosmological parameter estimation directly from CMB images.