Abstract: The dark matter sector remains completely unknown. It is therefore crucial to keep an open mind regarding its nature and possible interactions. Focusing on the case of Weakly Interacting Massive Particles, in this work we make this general philosophy more concrete by applying modern machine learning techniques to dark matter direct detection. We do this by encoding and decoding the graphical representation of background events in the XENONnT experiment with a convolutional variational autoencoder. We describe a methodology that utilizes the `anomaly score' derived from the reconstruction loss of the convolutional variational autoencoder as well as a pre-trained standard convolutional neural network, in a semi-supervised fashion. Indeed, we observe that optimum results are obtained only when both unsupervised and supervised anomaly scores are considered together. A data set that has a higher proportion of anomaly score is deemed anomalous and deserves further investigation. Contrary to classical analyses, in principle all information about the events is used, preventing unnecessary information loss. Lastly, we demonstrate the reach of learning-focused anomaly detection in this context by comparing results with classical inference, observing that, if tuned properly, these techniques have the potential to outperform likelihood-based methods.

Abstract: We study the impact of large lepton flavor asymmetries on the cosmic QCD transition. Scenarios of unequal lepton flavor asymmetries are observationally almost unconstrained and therefore open up a whole new parameter space for the cosmic QCD transition. We find that for large asymmetries, the formation of a Bose-Einstein condensate of pions can occur and identify the corresponding parameter space. In the vicinity of the QCD transition scale, we express the pressure in terms of a Taylor expansion with respect to the complete set of chemical potentials. The Taylor coefficients rely on input from lattice QCD calculations from the literature. The domain of applicability of this method is discussed.

Abstract: Picking arrival times of P and S phases is a fundamental and time‐consuming task for the routine processing of seismic data acquired by permanent and temporary networks. A large number of automatic pickers have been developed, but to perform well they often require the tuning of multiple parameters to adapt them to each dataset. Despite the great advance in techniques, some problems remain, such as the difficulty to accurately pick S waves and earthquake recordings with a low signal‐to‐noise ratio. Recently, phase pickers based on deep learning (DL) have shown great potential for event identification and arrival‐time picking. However, the general adoption of these methods for the routine processing of monitoring networks has been held back by factors such as the availability of well‐documented software, computational resources, and a gap in knowledge of these methods. In this study, we evaluate recent available DL pickers for earthquake data, comparing the performance of several neural network architectures. We test the selected pickers using three datasets with different characteristics. We found that the analyzed DL pickers (generalized phase detection, PhaseNet, and EQTransformer) perform well in the three tested cases. They are very efficient at ignoring large‐amplitude transient noise and at picking S waves, a task that is often difficult even for experienced analysts. Nevertheless, the performance of the analyzed DL pickers varies widely in terms of sensitivity and false discovery rate, with some pickers missing a significant percentage of true picks and others producing a large number of false positives. There are also variations in run time between DL pickers, with some of them requiring significant resources to process large datasets. In spite of these drawbacks, we show that DL pickers can be used efficiently to process large seismic datasets and obtain results comparable or better than current standard procedures.

Abstract: The Tibet AS gamma and LHAASO collaborations recently reported the observation of a gamma-ray diffuse emission with energy up to the PeV level from the Galactic plane.Aims. We discuss the relevance of non-uniform cosmic-ray transport scenarios and the implications of these results for cosmic-ray physics.Methods. We used the DRAGON and HERMES codes to build high-resolution maps and spectral distributions of that emission for several representative models under the condition that they reproduce a wide set of local cosmic-ray data up to 100 PeV.Results. We show that the energy spectra measured by Tibet AS gamma, LHAASO, ARGO-YBJ, and Fermi-LAT in several regions of interest in the sky can all be reasonably described in terms of the emission arising by the Galactic cosmic-ray “sea”. We also show that all our models are compatible with IceTop gamma-ray upper limits.Conclusions. We compare the predictions of conventional and space-dependent transport models with those data sets. Although the Fermi-LAT, ARGO-YBJ, and LHAASO preliminary data slightly favor this scenario, due to the still large experimental errors, the poorly known source spectral shape at the highest energies, the potential role of spatial fluctuations in the leptonic component, and a possible larger-than-expected contamination due to unresolved sources, a solid confirmation requires further investigations. We discuss which measurements will be most relevant in order to resolve the remaining degeneracy.

Abstract: Tuning is an essential requirement for the search of dark matter axions employing haloscopes since its mass is not known yet to the scientific community. At the present day, most haloscope tuning systems are based on mechanical devices which can lead to failures due to the complexity of the environment in which they are used. However, the electronic tuning making use of ferroelectric materials can provide a path that is less vulnerable to mechanical failures and thus complements and expands current tuning systems. In this work, we present and design a novel technique for using the ferroelectric Potassium Tantalate (KTaO3 or KTO) material as a tuning element in haloscopes based on coupled microwave cavities. In this line, the structures used in the Relic Axion Detector Exploratory Setup (RADES) group are based on several cavities that are connected by metallic irises, which act as interresonator coupling elements. In this article, we also show how to use these KTaO3 films as interresonator couplings between cavities, instead of inductive or capacitive metallic windows used in the past. These two techniques represent a crucial upgrade over the current systems employed in the dark matter axions community, achieving a tuning range of 2.23% which represents a major improvement as compared to previous works (<0.1%) for the same class of tuning systems. The theoretical and simulated results shown in this work demonstrate the interest of the novel techniques proposed for the incorporation of this kind of ferroelectric media in multicavity resonant haloscopes in the search for dark matter axions.