Recent studies using four-point correlations suggest a parity violation in the galaxy distribution, though the sensitivity of these detections is limited by the volume and accuracy of simulations used to model the noise properties of the galaxy distribution. In a recent paper, we introduced an unsupervised learning approach which offers an alternative method that significantly reduces or otherwise eliminates the dependence on mock catalogs, by learning parity violation directly from observational data. However, the convolutional neural network (CNN) model utilized by our previous unsupervised approach struggles to extend to more realistic scenarios where data is limited. We introduce a novel method, the neural field scattering transform (NFST), which enhances the wavelet scattering transform (WST) technique by adding trainable filters, parametrized as a neural field. The NFST is a general model that can flexibly and robustly capture information from a wide range of physical fields. We use the NFST for parity violation detection to demonstrate its ability to capture higher-order information. We compare the NFST’s performance at detecting parity violation in 2D mocks against WST and CNN benchmarks across varied training set sizes. We find the NFST can detect parity violation with $4\times$ less data than the CNN and $32\times$ less than the WST. Furthermore, in cases with limited data the NFST can detect parity violation with up to $6\sigma$ confidence, where the WST and CNN fail to make any detection. We identify that the added flexibility of the NFST, and particularly the ability to learn asymmetric filters, as well as the specific symmetries built into the NFST architecture, contribute to its improved performance over the benchmark models. We further demonstrate that the NFST’s detection capabilities are robust to several variations of the mock dataset, and we show that it is readily physically interpretable.

Close

PrevNext

View figure in article

• Large scale structure of the Universe