Deep learning has evolved as a strong and efficient framework that can be applied to a broad spectrum of complex learning problems which were difficult to solve using the traditional machine learning techniques in the past. The advancement of deep learning has been so radical that today it can surpass human-level performance. As a consequence, deep learning is being extensively used in most of the recent day-today applications. However, efficient deep learning systems can be jeopardised by using crafted adversarial samples, which may be imperceptible to the human eye, but can lead the model to misclassify the output. In recent times, different types of adversaries based on their threat model leverage these vulnerabilities to compromise a deep learning system where adversaries have high incentives. Hence, it is extremely important to provide robustness to deep learning algorithms against these adversaries. However, there are only a few strong countermeasures which can be used in all types of attack scenarios to design a robust deep learning system. Herein, the authors attempt to provide a detailed discussion on different types of adversarial attacks with various threat models and also elaborate on the efficiency and challenges of recent countermeasures against them. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

No abstract

No abstract

Cooper pairs in non-centrosymmetric superconductors can acquire finite centre-of-mass momentum in the presence of an external magnetic field. Recent theory predicts that such finite-momentum pairing can lead to an asymmetric critical current, where a dissipationless supercurrent can flow along one direction but not in the opposite one. Here we report the discovery of a giant Josephson diode effect in Josephson junctions formed from a type-II Dirac semimetal, NiTe2. A distinguishing feature is that the asymmetry in the critical current depends sensitively on the magnitude and direction of an applied magnetic field and achieves its maximum value when the magnetic field is perpendicular to the current and is of the order of just 10 mT. Moreover, the asymmetry changes sign several times with an increasing field. These characteristic features are accounted for by a model based on finite-momentum Cooper pairing that largely originates from the Zeeman shift of spin-helical topological surface states. The finite pairing momentum is further established, and its value determined, from the evolution of the interference pattern under an in-plane magnetic field. The observed giant magnitude of the asymmetry in critical current and the clear exposition of its underlying mechanism paves the way to build novel superconducting computing devices using the Josephson diode effect.

Electron Microscopy (EM) image (or volume) segmentation has become significantly important in recent years as an instrument for connectomics. This paper proposes a novel agglomerative framework for EM segmentation. In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron. Unlike existing agglomerative methods, the proposed context-aware algorithm divides superpixels (over-segmented regions) of different biological entities into different subsets and agglomerates them separately. In addition, this paper describes a “delayed” scheme for agglomerative clustering that postpones some of the merge decisions, pertaining to newly formed bodies, in order to generate a more confident boundary prediction. We report significant improvements attained by the proposed approach in segmentation accuracy over existing standard methods on 2D and 3D datasets.

There is considerable interest in van der Waals (vdW) materials as potential hosts for chiral skyrmionic spin textures. Of particular interest is the ferromagnetic, metallic compound Fe3GeTe2 (FGT), which has a comparatively high Curie temperature (150–220 K). Several recent studies have reported the observation of chiral Néel skyrmions in this compound, which is inconsistent with its presumed centrosymmetric structure. Here the observation of Néel type skyrmions in single crystals of FGT via Lorentz transmission electron microscopy (LTEM) is reported. It is shown from detailed X‐ray diffraction structure analysis that FGT lacks an inversion symmetry as a result of an asymmetric distribution of Fe vacancies. This vacancy‐induced breaking of the inversion symmetry of this compound is a surprising and novel observation and is a prerequisite for a Dzyaloshinskii–Moriya vector exchange interaction which accounts for the chiral Néel skyrmion phase. This phenomenon is likely to be common to many 2D vdW materials and suggests a path to the preparation of many such acentric compounds. Furthermore, it is found that the skyrmion size in FGT is strongly dependent on its thickness: the skyrmion size increases from ≈100 to ≈750 nm as the thickness of the lamella is increased from ≈90 nm to ≈2 µm. This extreme size tunability is a feature common to many low symmetry ferro‐ and ferri‐magnetic compounds.

No abstract

The electronic structure of sodium tungsten bronzes, Na x WO 3 , for full range of x is investigated by highresolution angle-resolved photoemission spectroscopy ͑HR-ARPES͒. The experimentally determined valenceband structure has been compared with the results of ab initio band-structure calculation. The HR-ARPES spectra taken in both the insulating and metallic phase of Na x WO 3 reveal the origin of metal-insulator transition ͑MIT͒ in the sodium tungsten bronze system. In the insulating Na x WO 3 , the near-E F states are localized due to the strong disorder caused by the random distribution of Na + ions in WO 3 lattice. While the presence of an impurity band ͑level͒ induced by Na doping is often invoked to explain the insulating state found at low concentrations, there is no signature of impurity band ͑level͒ found from our results. Due to disorder and Anderson localization effect, there is a long-range Coulomb interaction of conduction electrons; as a result, the system is insulating. In the metallic regime, the states near E F are populated and the Fermi level shifts upward rigidly with increasing electron doping ͑x͒. The volume of electronlike Fermi surface ͑FS͒ at the ⌫͑X͒ point gradually increases with increasing Na concentration due to W 5dt 2g band filling. A rigid shift of E F is found to give a qualitatively good description of the FS evolution.