Search

In this work, we present a novel approach to transform supervised classifiers into effective unsupervised anomaly detectors. The method we have developed, termed Discriminatory Detection of Distortions (DDD), enhances anomaly detection by training a discriminator model on both original and artificially modified datasets. We conducted a comprehensive evaluation of our models on the Dark Machines Anomaly Score Challenge channels and a search for 4-top quark events, demonstrating the effectiveness of our approach across various final states and beyond the Standard Model scenarios. We compare the performance of the DDD method with the Deep Robust One-Class Classification method (DROCC), which incorporates signals in the training process, and the Deep Support Vector Data Description (DeepSVDD) method, a well established and well performing method for anomaly detection. Results show that the effectiveness of each model varies by signal and channel, with DDD proving to be a very effective anomaly detector. We recommend the combined use of DeepSVDD and DDD for purely unsupervised applications, with the addition of flow models for improved performance when resources allow. Findings suggest that network architectures that excel in supervised contexts, such as the particle transformer with standard model interactions, also perform well as unsupervised anomaly detectors. We also show that with these methods, it is likely possible to recognize 4-top quark production as an anomaly without prior knowledge of the process. We argue that the Large Hadron Collider community can transform supervised classifiers into anomaly detectors to uncover potential new physical phenomena in each search.

The $U(1)_{L_\mu-L_\tau}$ extended Standard Model (SM) is anomaly free, and contains a massive $Z^\prime$ boson. The associated Higgs, which generates the $Z'$'s mass via spontaneous symmetry breaking (SSB) can mix with the SM Higgs. The new parameters relating to extra Higgs cannot be probed at the LHC with final states containing no more than $4$ leptons. Therefore, we use signatures with at least $6$ leptons to probe the parameter space through LHC experiments. Since SM predicts a lower cross section for a final state with at least $6$ leptons than is currently visible at the LHC, the background is negligible, thereby making this channel an extremely sensitive probe. We find that in a limited region of the parameter space this channel can strongly constrain the associated $U(1)_{L_\mu-L_\tau}$ coupling constant, even more so than for a $4$ lepton or fewer final state., Comment: 5 pages, 3 figures. Minor changes based on comments received

We classify the pairing symmetries of three-dimensional superconductivity in the zinc-blende structure which can support an electronic environment to host unconventional high temperature superconductivity, and calculate the pairing symmetry in the presence of strong electron-electron correlation by the slave boson mean-field approach. We find that the $d_{2z^2-x^2-y^2} \pm id_{x^2-y^2}$ pairing state, a three dimensional analogy of the $d\pm id$ pairing in a two dimensional square lattice, is ubiquitously favored near half filling upon hole doping in both single-orbital and three-orbital models. However, unlike the two dimensional counterpart, the Bogoliubov quasiparticle spectrum of the three dimensional state upholds the full $T_d$ point group symmetry and encompasses point nodes along certain high symmetric lines., Comment: 9pages,6 figures and 7 tabulations

Neutral particles capable of travelling cosmic distances from a source to detectors on Earth are limited to photons and neutrinos. Examination of the Dark Matter annihilation/decay spectra for these particles reveals the presence of continuum spectra (e.g. due to fragmentation and W or Z decay) and peaks (due to direct annihilations/decays). However, when one explores extensions of the Standard Model (BSM), unexplored spectra emerge that differ significantly from those of the Standard Model (SM) for both neutrinos and photons. In this paper, we argue for the inclusion of important spectra that include peaks as well as previously largely unexplored entities such as boxes and combinations of box, peak and continuum decay spectra., Comment: 7 pages, 2 figures

We investigate the degenerate fractional Schr\"{o}dinger-Kirchhoff-Poisson equation in $\mathbb{R}^3$ with critical nonlinearity and electromagnetic fields $\varepsilon^{2s} M([u]_{s,A}^2)(-\Delta)_{A}^su + V(x)u + \phi u = k(x)|u|^{r-2}u + \left(\mathcal{I}_\mu*|u|^{2_s^\sharp}\right)|u|^{2_s^\sharp-2}u$ and $(-\Delta)^t\phi = u^2,$ where $\varepsilon > 0$ is a parameter, $3/4

Cross-domain recommendation (CDR) is an effective way to alleviate the data sparsity problem. Content-based CDR is one of the most promising branches since most kinds of products can be described by a piece of text, especially when cold-start users or items have few interactions. However, two vital issues are still under-explored: (1) From the content modeling perspective, sufficient long-text descriptions are usually scarce in a real recommender system, more often the light-weight textual features, such as a few keywords or tags, are more accessible, which is improperly modeled by existing methods. (2) From the CDR perspective, not all inter-domain interests are helpful to infer intra-domain interests. Caused by domain-specific features, there are part of signals benefiting for recommendation in the source domain but harmful for that in the target domain. Therefore, how to distill useful interests is crucial. To tackle the above two problems, we propose a metapath and multi-interest aggregated graph neural network (M2GNN). Specifically, to model the tag-based contents, we construct a heterogeneous information network to hold the semantic relatedness between users, items, and tags in all domains. The metapath schema is predefined according to domain-specific knowledge, with one metapath for one domain. User representations are learned by GNN with a hierarchical aggregation framework, where the intra-metapath aggregation firstly filters out trivial tags and the inter-metapath aggregation further filters out useless interests. Offline experiments and online A/B tests demonstrate that M2GNN achieves significant improvements over the state-of-the-art methods and current industrial recommender system in Dianping, respectively. Further analysis shows that M2GNN offers an interpretable recommendation.

A major task in particle physics is the measurement of rare signal processes. These measurements are highly dependent on the classification accuracy of these events in relation to the huge background of other Standard Model processes. Reducing the background by a few tens of percent with the same signal efficiency can already increase the sensitivity considerably. This work demonstrates the importance of incorporating physical information into deep learning-based event selection. The paper includes this information into different methods for classifying events, in particular Boosted Decision Trees, Transformer Architectures (Particle Transformer) and Graph Neural Networks (Particle Net). In addition to the physical information previously proposed for jet tagging, we add particle measures for energy-dependent particle-particle interaction strengths as predicted by the leading order interactions of the Standard Model (SM). We find that the integration of physical information into the attention matrix (transformers) or edges (graphs) notably improves background rejection by $10\%$ to $40\%$ over baseline models (a graph network), with about $10\%$ of this improvement directly attributable to what we call the SM interaction matrix. In a simplified statistical analysis, we find that such architectures can improve the significance of signals by a significant factor compared to a graph network (our base model)., Comment: Version 2 of this paper supersedes arXiv:2211.05143v1. 26 pages, 5 figures

Despite high global prevalence of hepatic steatosis, no automated diagnostics demonstrated generalizability in detecting steatosis on multiple international datasets. Traditionally, hepatic steatosis detection relies on clinicians selecting the region of interest (ROI) on computed tomography (CT) to measure liver attenuation. ROI selection demands time and expertise, and therefore is not routinely performed in populations. To automate the process, we validated an existing artificial intelligence (AI) system for 3D liver segmentation and used it to purpose a novel method: AI-ROI, which could automatically select the ROI for attenuation measurements. AI segmentation and AI-ROI method were evaluated on 1,014 non-contrast enhanced chest CT images from eight international datasets: LIDC-IDRI, NSCLC-Lung1, RIDER, VESSEL12, RICORD-1A, RICORD-1B, COVID-19-Italy, and COVID-19-China. AI segmentation achieved a mean dice coefficient of 0.957. Attenuations measured by AI-ROI showed no significant differences (p = 0.545) and a reduction of 71% time compared to expert measurements. The area under the curve (AUC) of the steatosis classification of AI-ROI is 0.921 (95% CI: 0.883 - 0.959). If performed as a routine screening method, our AI protocol could potentially allow early non-invasive, non-pharmacological preventative interventions for hepatic steatosis. 1,014 expert-annotated liver segmentations of patients with hepatic steatosis annotations can be downloaded here: https://drive.google.com/drive/folders/1-g_zJeAaZXYXGqL1OeF6pUjr6KB0igJX.

We show that there are diversified Ruderman-Kittel-Kasuya-Yosida (RKKY) interactions between magnetic impurities, mediated by itinerant electrons, in a centrosymmetric crystal respecting a nonsymmorphic space group. We take the $P4/nmm$ space group as an example. We demonstrate that the different type of interactions, including the Heisenberg-type, the Dzyaloshinskii-Moriya (DM)-type, the Ising-type and the anisotropic interactions, can appear in accordance with the positions of the impurities in the real space. Their strengths strongly depend on the location of the itinerant electrons in the reciprocal space. The diversity stems from the position-dependent site groups and the momentum-dependent electronic structures guaranteed by the nonsymmorphic symmetries. Our study unveils the role of the nonsymmorphic symmetries in affecting magnetism, and suggests that the nonsymmorphic crystals can be promising platforms to design magnetic interactions., Comment: 11pages, 5 figures

We show that topological superconductivity can be generally induced in a magnetic metal through the superconducting proximity effect. In this case, the topological superconductivity originates from band degeneracies controlled by crystalline symmetries. We demonstrate this general scheme in a model with a 4-fold band degeneracy protected by the space group P4=nmm. We show that first order or second-order topological superconductivity can be realized in the presence of a ferromagnetic order or an antiferromagnetic order respectively. We derive the corresponding topological invariants and analyze Majorana modes. Our study provides a general method to realize topological superconductivity and help to identify new platforms to generate Majorana zero modes.

Model-independent search strategies have been increasingly proposed in recent years because on the one hand there has been no clear signal for new physics and on the other hand there is a lack of a highly probable and parameter-free extension of the standard model. For these reasons, there is no simple search target so far. In this work, we try to take a new direction and ask the question: bearing in mind that we have a large number of new physics theories that go beyond the Standard Model and may contain a grain of truth, can we improve our search strategy for unknown signals by using them "in combination"? In particular, we show that a signal hypothesis based on a large, intermingled set of many different theoretical signal models can be a superior approach to find an unknown BSM signal. Applied to a recent data challenge, we show that "mixture-of-theories training" outperforms strategies that optimize signal regions with a single BSM model as well as most unsupervised strategies. Applications of this work include anomaly detection and the definition of signal regions in the search for signals of new physics., Comment: 40 pages, 10 figures

Neutrino telescope experiments are rapidly becoming more competitive in indirect detection searches for dark matter. Neutrino signals arising from dark matter annihilations are typically assumed to originate from the hadronisation and decay of Standard Model particles. Here we showcase a supersymmetric model, the BLSSMIS, that can simultaneously obey current experimental limits while still providing a potentially observable non-standard neutrino spectrum from dark matter annihilation., Comment: 10 pages, 6 figures

Purpose: Automatic methods are required for the early detection of hepatic steatosis to avoid progression to cirrhosis and cancer. Here, we developed a fully automated deep learning pipeline to quantify hepatic steatosis on non-contrast enhanced chest computed tomography (CT) scans. Materials and Methods: We developed and evaluated our pipeline on chest CT images of 1,431 randomly selected National Lung Screening Trial (NLST) participants. A dataset of 451 CT scans with volumetric liver segmentations of expert readers was used for training a deep learning model. For testing, in an independent dataset of 980 CT scans hepatic attenuation was manually measured by an expert reader on three cross-sectional images at different hepatic levels by selecting three circular regions of interest. Additionally, 100 randomly selected cases of the test set were volumetrically segmented by expert readers. Hepatic steatosis on the test set was defined as mean hepatic attenuation of < 40 Hounsfield unit. Spearman correlation was conducted to analyze liver fat quantification accuracy and the Cohen's Kappa coefficient was calculated for hepatic steatosis prediction reliability. Results: Our pipeline demonstrated strong performance and achieved a mean dice score of 0.970 for the volumetric liver segmentation. The spearman correlation of the liver fat quantification was 0.954 (P <0.0001) between the automated and expert reader measurements. The cohen's kappa coefficient was 0.875 for automatic assessment of hepatic steatosis. Conclusion: We developed a fully automatic deep learning-based pipeline for the assessment of hepatic steatosis in chest CT images. With the fast and cheap screening of hepatic steatosis, our pipeline has the potential to help initiate preventive measures to avoid progression to cirrhosis and cancer.

Purpose: Identifying intravenous (IV) contrast use within CT scans is a key component of data curation for model development and testing. Currently, IV contrast is poorly documented in imaging metadata and necessitates manual correction and annotation by clinician experts, presenting a major barrier to imaging analyses and algorithm deployment. We sought to develop and validate a convolutional neural network (CNN)-based deep learning (DL) platform to identify IV contrast within CT scans. Methods: For model development and evaluation, we used independent datasets of CT scans of head, neck (HN) and lung cancer patients, totaling 133,480 axial 2D scan slices from 1,979 CT scans manually annotated for contrast presence by clinical experts. Five different DL models were adopted and trained in HN training datasets for slice-level contrast detection. Model performances were evaluated on a hold-out set and on an independent validation set from another institution. DL models was then fine-tuned on chest CT data and externally validated on a separate chest CT dataset. Results: Initial DICOM metadata tags for IV contrast were missing or erroneous in 1,496 scans (75.6%). The EfficientNetB4-based model showed the best overall detection performance. For HN scans, AUC was 0.996 in the internal validation set (n = 216) and 1.0 in the external validation set (n = 595). The fine-tuned model on chest CTs yielded an AUC: 1.0 for the internal validation set (n = 53), and AUC: 0.980 for the external validation set (n = 402). Conclusion: The DL model could accurately detect IV contrast in both HN and chest CT scans with near-perfect performance.

Intrinsic topological superconductors have protected gapless Majorana modes, bound and/or propagating, at the natural boundaries of the sample, without requiring field, defect, or heterostructure. We establish the complete classification/construction of intrinsic topological superconductors jointly protected by point-group and time-reversal symmetries in three dimensions. This is obtained from enumerating distinct ways for stacking $n$th-order irreducible building blocks, minimal anomalous surface states of $n$th-order topological superconductors. Particularly, our method provides a unified description of possible surface anomalies away from high-symmetry points/lines in terms of the homotopy group of the surface mass field., Comment: 43 pages, 9 figures, 6 tables

Extending the Standard Model (SM) by a $U(1)_{L_\mu-L_\tau}$ group gives potentially significant new contributions to $g_\mu-2$, allows the construction of realistic neutrino mass matrices, incorporates lepton universality violation, and offers an anomaly-free mediator for a Dark Matter (DM) sector. In a recent analysis we showed that published LHC searches are not very sensitive to this model. Here we apply several Machine Learning (ML) algorithms in order to distinguish this model from the SM using simulated LHC data. In particular, we optimize the $3\mu$-signal, which has a considerably larger cross section than the $4\mu$-signal. Furthermore, since the $2$-muon plus missing $E_T$ final state gets contributions from diagrams involving DM particles, we optimize it as well. We find greatly improved sensitivity, which already for $36$ fb$^{-1}$ of data exceeds the combination of published LHC and non-LHC results. We also emphasize the usefulness of Boosted Decision Trees which, unlike Neural Networks, easily allow to extract additional information from the data which directly connect to the theoretical model through feature importance. The same scheme could be used to analyze other models., Comment: 39 pages, 10 figures

For three-dimensional metals, Landau levels disperse as a function of the magnetic field and the momentum wavenumber parallel to the field. In this two-dimensional parameter space, it is shown that two conically-dispersing Landau levels can touch at a diabolical point -- a Landau-Dirac point. The conditions giving rise to Landau-Dirac points are shown to be magnetic breakdown (field-induced quantum tunneling) and certain crystallographic spacetime symmetry. Both conditions are realizable in topological nodal-line metals, as we exemplify with CaP$_3$. A Landau-Dirac point reveals itself in anomalous batman-like peaks in the magnetoresistance, as well as in the onset of optical absorption linearly evolving to zero frequency as a function of the field magnitude/orientation.

We introduce an $A_\infty$ map from the cubical chain complex of the based loop space of Lagrangian submanifolds with Legendrian boundary in a Liouville Manifold $C_{*}(\Omega_{L} \mathcal{L}\mathit{ag})$ to wrapped Floer cohomology of Lagrangian submanifold $\mathcal{CW}^{-*}(L,L)$. In the case of a cotangent bundle and a Lagrangian co-fiber, the composition of our map with a previously constructed map from $\mathcal{CW}^*(L,L) \to C_{*}(\Omega_q Q) $ shows that this map is split surjective., Comment: arXiv admin note: text overlap with arXiv:0907.5606 by other authors