Flagging of Radio Frequency Interference (RFI) is an increasingly important challenge in radio astronomy. We present R-Net, a deep convolutional ResNet architecture that significantly outperforms existing algorithms -- including the default MeerKAT RFI flagger, and deep U-Net architectures -- across all metrics including AUC, F1-score and MCC. We demonstrate the robustness of this improvement on both single dish and interferometric simulations and, using transfer learning, on real data. Our R-Net model's precision is approximately $90\%$ better than the current MeerKAT flagger at $80\%$ recall and has a 35\% higher F1-score with no additional performance cost. We further highlight the effectiveness of transfer learning from a model initially trained on simulated MeerKAT data and fine-tuned on real, human-flagged, KAT-7 data. Despite the wide differences in the nature of the two telescope arrays, the model achieves an AUC of 0.91, while the best model without transfer learning only reaches an AUC of 0.67. We consider the use of phase information in our models but find that without calibration the phase adds almost no extra information relative to amplitude data only. Our results strongly suggest that deep learning on simulations, boosted by transfer learning on real data, will likely play a key role in the future of RFI flagging of radio astronomy data.

In this work, we propose a new method to inpaint the CMB signal in regions masked out following a point source extraction process. We adopt a modified Generative Adversarial Network (GAN) and compare different combinations of internal (hyper-)parameters and training strategies. We study the performance using a suitable $\mathcal{C}_r$ variable in order to estimate the performance regarding the CMB power spectrum recovery. We consider a test set where one point source is masked out in each sky patch with a 1.83 $\times$ 1.83 squared degree extension, which, in our gridding, corresponds to 64 $\times$ 64 pixels. The GAN is optimized for estimating performance on Planck 2018 total intensity simulations. The training makes the GAN effective in reconstructing a masking corresponding to about 1500 pixels with $1\%$ error down to angular scales corresponding to about 5 arcminutes.

The clustering of local extrema including peaks and troughs will be exploited to assess Gaussianity, asymmetry and the footprint of cosmic strings network on the CMB random field observed by {\it Planck} satellite. The number density of local extrema reveals some non-resolved shot noise in {\it Planck} maps. The \texttt{SEVEM} data has maximum number density of peaks, $n_{pk}$, and troughs, $n_{tr}$, compared to other observed maps. The cumulative of $n_{pk}$ and $n_{tr}$ above and below a threshold, $\vartheta$, for all {\it Planck} maps except for the 100GHz band are compatible with the Gaussian random field. The unweighted Two-Point Correlation Function (TPCF), $\Psi(\theta;\vartheta)$, of the local extrema illustrates significant non-Gaussianity for angular separation $\theta\le 15'$ for all available thresholds. Our results show that to put the feasible constraint on the amplitude of the mass function based on the value of $\Psi$ around the {\it Doppler peak} ($\theta\approx 70'-75'$), we should consider $\vartheta\gtrsim+1.0$. The scale independent bias factors for peak above a threshold for large separation angle and high threshold level are in agreement with that of expected for a pure Gaussian CMB. Unweighted TPCF of local extrema demonstrates a level of rejecting Gaussian hypothesis in \texttt{SMICA}. Genus topology also confirms the Gaussian hypothesis for different component separation maps. Tessellating CMB map with disk of size $6^{\circ}$ based on $n_{pk}$ and $\Psi_{pk-pk}$ demonstrate statistical symmetry in {\it Planck} maps. Combining all maps and applying the $\Psi_{pk-pk}$ puts the upper bound on the cosmic string's tension: $G\mu^{(up)} \lesssim 5.00\times 10^{-7}$.

Anomaly detection is challenging, especially for large datasets in high dimensions. Here we explore a general anomaly detection framework based on dimensionality reduction and unsupervised clustering. We release DRAMA, a general python package that implements the general framework with a wide range of built-in options. We test DRAMA on a wide variety of simulated and real datasets, in up to 3000 dimensions, and find it robust and highly competitive with commonly-used anomaly detection algorithms, especially in high dimensions. The flexibility of the DRAMA framework allows for significant optimization once some examples of anomalies are available, making it ideal for online anomaly detection, active learning and highly unbalanced datasets.

We explore the potential of the B-mode anisotropies of the Cosmic Microwave Background radiation (CMB) to constrain the shape of the primordial tensor power spectrum in a model-independent way. We expand possible perturbations to the power-law primordial tensor spectrum (predicted by the simplest single-field slow-roll inflationary models) using various sets of localized and nonlocalized basis functions and construct the Fisher matrix for their amplitudes. The eigen-analysis of the Fisher matrix would then yield a hierarchy of uncorrelated perturbation patterns (called tensor eigenmodes or TeMs) which are rank-ordered according to their measurability by data. We find that the first three TeMs are expected to be constrainable within a few percent by the next generation of B-mode experiments. We discuss how the method can be iteratively used to reconstruct the observable part of any general deviation from the fiducial power spectrum.

Point source detection at low signal-to-noise is challenging for astronomical surveys, particularly in radio interferometry images where the noise is correlated. Machine learning is a promising solution, allowing the development of algorithms tailored to specific telescope arrays and science cases. We present DeepSource - a deep learning solution - that uses convolutional neural networks to achieve these goals. DeepSource enhances the Signal-to-Noise Ratio (SNR) of the original map and then uses dynamic blob detection to detect sources. Trained and tested on two sets of 500 simulated 1 deg x 1 deg MeerKAT images with a total of 300,000 sources, DeepSource is essentially perfect in both purity and completeness down to SNR = 4 and outperforms PyBDSF in all metrics. For uniformly-weighted images it achieves a Purity x Completeness (PC) score at SNR = 3 of 0.73, compared to 0.31 for the best PyBDSF model. For natural-weighting we find a smaller improvement of ~40% in the PC score at SNR = 3. If instead we ask where either of the purity or completeness first drop to 90%, we find that DeepSource reaches this value at SNR = 3.6 compared to the 4.3 of PyBDSF (natural-weighting). A key advantage of DeepSource is that it can learn to optimally trade off purity and completeness for any science case under consideration. Our results show that deep learning is a promising approach to point source detection in astronomical images.

We explore the use of random forest and gradient boosting, two powerful tree-based machine learning algorithms, for the detection of cosmic strings in maps of the cosmic microwave background (CMB), through their unique Gott-Kaiser-Stebbins effect on the temperature anisotropies.The information in the maps is compressed into feature vectors before being passed to the learning units. The feature vectors contain various statistical measures of processed CMB maps that boost the cosmic string detectability. Our proposed classifiers, after training, give results improved over or similar to the claimed detectability levels of the existing methods for string tension, $G\mu$. They can make $3\sigma$ detection of strings with $G\mu \gtrsim 2.1\times 10^{-10}$ for noise-free, $0.9'$-resolution CMB observations. The minimum detectable tension increases to $G\mu \gtrsim 3.0\times 10^{-8}$ for a more realistic, CMB S4-like (II) strategy, still a significant improvement over the previous results.

We propose a multi-scale edge-detection algorithm to search for the Gott-Kaiser-Stebbins imprints of a cosmic string (CS) network on the Cosmic Microwave Background (CMB) anisotropies. Curvelet decomposition and extended Canny algorithm are used to enhance the string detectability. Various statistical tools are then applied to quantify the deviation of CMB maps having a cosmic string contribution with respect to pure Gaussian anisotropies of inflationary origin. These statistical measures include the one-point probability density function, the weighted two-point correlation function (TPCF) of the anisotropies, the unweighted TPCF of the peaks and of the up-crossing map, as well as their cross-correlation. We use this algorithm on a hundred of simulated Nambu-Goto CMB flat sky maps, covering approximately $10\%$ of the sky, and for different string tensions $G\mu$. On noiseless sky maps with an angular resolution of $0.9'$, we show that our pipeline detects CSs with $G\mu$ as low as $G\mu\gtrsim 4.3\times 10^{-10}$. At the same resolution, but with a noise level typical to a CMB-S4 phase II experiment, the detection threshold would be to $G\mu\gtrsim 1.2 \times 10^{-7}$.

In this work we study the imprints of a primordial cosmic string on inflationary power spectrum. Cosmic string induces two distinct contributions on curvature perturbations power spectrum. The first type of correction respects the translation invariance while violating isotropy. This generates quadrupolar statistical anisotropy in CMB maps which is constrained by the Planck data. The second contribution breaks both homogeneity and isotropy, generating a dipolar power asymmetry in variance of temperature fluctuations with its amplitude falling on small scales. We show that the strongest constraint on the tension of string is obtained from the quadrupolar anisotropy and argue that the mass scale of underlying theory responsible for the formation of string can not be much higher than the GUT scale. The predictions of string for the diagonal and off-diagonal components of CMB angular power spectrum are presented.