9682 research outputs found
Sort by
Mitigation of plasma-wall interactions with low-Z powders in DIII-D high confinement plasmas
No full textExperiments with low-Z powder injection in DIII-D high confinement discharges demonstrated increased divertor dissipation and detachment while maintaining good core energy confinement. Lithium (Li), boron (B), and boron nitride (BN) powders were injected in H-mode plasmas (Ip =1 MA, Bt =2 T, PNB =6 MW, ⟨ne⟩ = 3.6 − 5.0 · 1019 m−3) into the upper small-angle slot (SAS) divertor for 2-s intervals at constant rates of 3-204 mg/s. The multi-species BN powders at a rate of 54 mg/s showed the most substantial increase in divertor neutral compression by more than an order of magnitude and lasting detachment with minor degradation of the stored magnetic energy Wmhd by 5%. Rates of 204 mg/s of boron nitride powder further reduce ELM-fluxes on the divertor but also cause a drop in confinement performance by 24% due to the onset of an n = 2 tearing mode. The application of powders also showed a substantial improvement of wall conditions manifesting in reduced wall fueling source and intrinsic carbon and oxygen content in response to the cumulative injection of non-recycling materials. The results suggest that low-Z powder injection, including mixed element compounds, is a promising new core-edge compatible technique that simultaneously enables divertor detachment and improves wall conditions during high confinement operation
Search for low-energy signals from fast radio bursts with the Borexino detector
No full textAbstractThe search for neutrino events in correlation with 42 most intense fast radio bursts (FRBs) has been performed using the Borexino dataset from 05/2007 to 06/2021. We have searched for signals with visible energies above 250 keV within a time window of ± 1000 s corresponding to detection time of a particular FRB. We also applied an alternative approach based on searching for specific shapes of neutrino-electron scattering spectra in the full exposure data of the Borexino detector. In particular, two incoming neutrino spectra were considered: the monoenergetic line and the spectrum expected from supernovae. The same spectra were considered for electron antineutrinos detected through inverse beta-decay reaction. No statistically significant excess over the background was observed. As a result, the strongest upper limits on FRB-associated neutrino fluences of all flavors have been obtained in the 0.5–50 MeV neutrino energy range.</jats:p
XNAS: A Regressive/Progressive NAS for Deep Learning
No full textDeep learning has achieved great and broad breakthroughs in many real-world applications. In particular, the task of training the network parameters has been masterly handled by back-propagation learning. However, the pursuit on optimal network structures remains largely an art of trial and error. This prompts some urgency to explore an architecture engineering process, collectively known as Neural Architecture Search (NAS). In general, NAS is a design software system for automating the search of effective neural architecture. This article proposes an X-learning NAS (XNAS) to automatically train a network’s structure and parameters. Our theoretical footing is built upon the subspace and correlation analyses between the input layer, hidden layer, and output layer. The design strategy hinges upon the underlying principle that the network should be coerced to learn how to structurally improve the input/output correlation successively (i.e., layer by layer). It embraces both Progressive NAS (PNAS) and Regressive NAS (RNAS). For unsupervised RNAS, Principal Component Analysis (PCA) is a classic tool for subspace analyses. By further incorporating teacher’s guidance, PCA can be extended to Regression Component Analysis (RCA) to facilitate supervised NAS design. This allows the machine to extract components most critical to the targeted learning objective. We shall further extend the subspace analysis from multi-layer perceptrons to convolutional neural networks, via introduction of Convolutional-PCA (CPCA) or, more simply, Deep-PCA (DPCA). The supervised variant of DPCA will be named Deep-RCA (DRCA). The subspace analyses allow us to compute optimal eigenvectors (respectively, eigen-filters) and principal components (respectively, eigen-channels) for optimal NAS design of multi-layer perceptrons (respectively, convolutional neural networks). Based on the theoretical analysis, an X-learning paradigm is developed to jointly learn the structure and parameters of learning models. The objective is to reduce the network complexity while retaining (and sometimes improving) the performance. With carefully pre-selected baseline models, X-learning has shown great successes in numerous classification-type and/or regression-type applications. We have applied X-learning to the ImageNet datasets for classification and DIV2K for image enhancements. By applying X-learning to two types of baseline models, MobileNet and ResNet, both the low-power and high-performance application categories can be supported. Our simulations confirm that X-learning is by and large very competitive relative to the state-of-the-art approaches
Observation of Reentrant Correlated Insulators and Interaction-Driven Fermi-Surface Reconstructions at One Magnetic Flux Quantum per Moiré Unit Cell in Magic-Angle Twisted Bilayer Graphene
No full textThe discovery of flat bands with non-trivial band topology in magic angle twisted bilayer graphene (MATBG) has provided a unique platform to study strongly correlated phenomena including superconductivity, correlated insulators, Chern insulators and magnetism. A funda mental feature of the MATBG, so far unexplored, is its high magnetic field Hofstadter spectrum. Here we report on a detailed magneto-transport study of a MATBG device in external magnetic fields of up to B = 31 T, corresponding to one magnetic flux quantum per moiré unit cell Φ0. At Φ0, we observe a re-entrant correlated insulator at a flat band filling factor of ν = +2, and inter action-driven Fermi surface reconstructions at other fillings, which are identified by new sets of Landau levels originating from these. These experimental observations are supplemented by the oretical work that predicts a new set of 8 well-isolated flat bands at Φ0 , of comparable band width
but with different topology than in zero field. Overall, our magneto-transport data reveals a qual itatively new Hofstadter spectrum in MATBG, which arises due to the strong electronic correla tions in the re-entrant flat bands
Fermions in AdS and Gross-Neveu BCFT
No full textAbstract We study the boundary critical behavior of conformal field theories of interacting fermions in the Gross-Neveu universality class. By a Weyl transformation, the problem can be studied by placing the CFT in an anti de Sitter space background. After reviewing some aspects of free fermion theories in AdS, we use both large N methods and the epsilon expansion near 2 and 4 dimensions to study the conformal boundary conditions in the Gross-Neveu CFT. At large N and general dimension d, we find three distinct boundary conformal phases. Near four dimensions, where the CFT is described by the Wilson-Fisher fixed point of the Gross-Neveu-Yukawa model, two of these phases correspond respectively to the choice of Neumann or Dirichlet boundary condition on the scalar field, while the third one corresponds to the case where the bulk scalar field acquires a classical expectation value. One may flow between these boundary critical points by suitable relevant boundary deformations. We compute the AdS free energy on each of them, and verify that its value is consistent with the boundary version of the F-theorem. We also compute some of the BCFT observables in these theories, including bulk two-point functions of scalar and fermions, and four-point functions of boundary fermions.</jats:p
Learning Probabilistic Protein-DNA Recognition Codes from DNA-Binding Specificities Using Structural Mappings
No full textKnowledge of how proteins interact with DNA is essential for understanding gene regulation. Although DNA-binding specificities for thousands of transcription factors (TFs) have been determined, the specific amino acid–base interactions comprising their structural interfaces are largely unknown. This lack of resolution hampers attempts to leverage these data in
order to predict specificities for uncharacterized TFs or TFs mutated in disease. Here we introduce recognition code learning
via automated mapping of protein–DNA structural interfaces (rCLAMPS), a probabilistic approach that uses DNA-binding
specificities for TFs from the same structural family to simultaneously infer both which nucleotide positions are contacted
by particular amino acids within the TF as well as a recognition code that relates each base-contacting amino acid to nucleotide preferences at the DNA positions it contacts. We apply rCLAMPS to homeodomains, the second largest family of TFs
in metazoans and show that it learns a highly effective recognition code that can predict de novo DNA-binding specificities
for TFs. Furthermore, we show that the inferred amino acid–nucleotide contacts reveal whether and how nucleotide preferences at individual binding site positions are altered by mutations within TFs. Our approach is an important step toward
automatically uncovering the determinants of protein–DNA specificity from large compendia of DNA-binding specificities
and inferring the altered functionalities of TFs mutated in disease
Influence of Orbital Character on the Ground State Electronic Properties in the van Der Waals Transition Metal Iodides VI3 and CrI3
No full textTwo-dimensional van der Waals magnetic semiconductors display emergent chemical and physical properties and hold promise for novel optical, electronic and magnetic “few-layers” functionalities. Transition-metal iodides such as CrI3 and VI3 are relevant for future electronic and spintronic applications; however, detailed experimental information on their ground state electronic properties is lacking often due to their challenging chemical environment. By combining X-ray electron spectroscopies and first-principles calculations, we report a complete determination of CrI3 and VI3 electronic ground states. We show that the transition metal-induced orbital filling drives the stabilization of distinct electronic phases: a wide bandgap in CrI3 and a Mott insulating state in VI3. Comparison of surface-sensitive (angular-resolved photoemission spectroscopy) and bulk-sensitive (X-ray absorption spectroscopy) measurements in VI3 reveals a surface-only V2+ oxidation state, suggesting that ground state electronic properties are strongly influenced by dimensionality effects. Our results have direct implications in band engineering and layer-dependent properties of two-dimensional systems
Reentrant Correlated Insulators in Twisted Bilayer Graphene at 25T 2π Flux
No full textTwisted bilayer graphene (TBG) is remarkable for its topological flat bands, which drive strongly interacting physics at integer fillings, and its simple theoretical description facilitated by the Bistritzer-MacDonald Hamiltonian, a continuum model coupling two Dirac fermions. Because of the large moiré unit cell, TBG offers the unprecedented opportunity to observe reentrant Hofstadter phases in laboratory-strength magnetic fields near 25 T. This Letter is devoted to magic angle TBG at 2π flux where the magnetic translation group commutes. We use a newly developed gauge-invariant formalism to determine the exact single-particle band structure and topology. We find that the characteristic TBG flat bands reemerge at
2π flux, but, due to the magnetic field breaking C2zT, they split and acquire Chern number ±1. We show that reentrant correlated insulating states appear at 2π flux driven by the Coulomb interaction at integer fillings, and we predict the characteristic Landau fans from their excitation spectrum