As optical frequency nanoantennas, reduced-symmetry plasmonic nanoparticles have light-scattering properties that depend strongly on geometry, orientation, and variations in dielectric environment. Here we investigate how these factors influence the spectral angular dependence of light scattered by Au nanocups. A simple substrate causes axial, electric dipole mode nanocup to deviate substantially from its characteristic cos(2) θ free space scattering profile, while transverse, magnetic...

Under the second-order degenerate perturbation theory, we show that physics of $N$ particles with arbitrary spin confined in a one dimensional trap strongly interacting regime can be described by super-exchange interaction. An effective spin-chain Hamiltonian (non-translational-invariant Sutherland model) constructed from this procedure. For spin-1/2 particles, model reduces to non-translational-invariant Heisenberg model, where transition between anti-ferromagnetic (AFM) and ferromagnetic...

The authors construct and develop an optimization scheme to train a deep convolutional neural network represent many-body wave function. paper explores its performance by applying the find ground state of SU(N) spin-chain Hamiltonian using variational quantum Monte Carlo.

A large number of Internet-of-Things (IoT) devices will soon populate our physical environments. Yet, IoT devices' reliance on mobile applications and voice-only assistants as the primary interface limits their scalability expressiveness. Building off classic 'Put-That-There' system, we contribute an exploration design space voice + gesture interaction with spatially-distributed devices. Our decomposes users' commands into two components—selection interaction. We articulate how permutations...

We show that the wave function of a one dimensional spinor gas with contact $s$-wave interaction, either bosonic or fermionic, can be mapped to direct product spinless Fermi short-range $p$-wave interaction and spin system governed by parity projection operators. Applying this mapping strongly interacting gases, we obtain generalized chain model captures both static dynamics properties system. Using model, investigate breathing mode frequency quench harmonically trapped gases.

Flight itinerary ranking is critical for Online Travel Agencies (OTAs) since more and customers book flights online. Currently, most OTAs still adopt rule-based strategies. However, methods are not able to model context-aware information user preferences. To this end, a novel Personalized Ranking Network (PFRN) proposed in paper. In PFRN, Listwise Feature Encoding (LFE) structure capture global mutual influences among inputs. Besides, we utilize behaviors of both individual group users...

Dynamical fermionization refers to the phenomenon in Tonks-Girardeau gases where, upon release from harmonic confinement, gases' momentum density profile evolves asymptotically that of an ideal Fermi gas initial trap. This has been demonstrated theoretically hardcore and anyonic was recently experimentally observed a strongly interacting Bose gas. We extend this study one-dimensional spinor arbitrary spin regime analytically prove total distribution after trap is turned off approaches...

Evaluation of policies in recommender systems typically involves A/B testing using live experiments on real users to assess a new policy's impact relevant metrics. This ``gold standard'' comes at high cost, however, terms cycle time, user and potential retention. In developing for ``onboarding'' users, these costs can be especially problematic, since on-boarding occurs only once. this work, we describe simulation methodology used augment (and reduce) the use experiments. We illustrate its...

Deep neural networks have been shown as a potentially powerful ansatz in variational Monte Carlo for solving quantum many-body problems. We propose two improvements this direction. The first is graph (GNA), which wavefunction universal to arbitrary geometry. GNA results accurate ground-state energies on 2D Kagome lattices, triangular and randomly connected graphs. Secondly, we design distributed workflow multiple accelerators scale up the computation. compute lattices with sizes 432 sites...

The one-body density matrix (OBDM) of a strongly interacting spinor quantum gas in one dimension can be written as summation products spatial and spin parts. We find that there is remarkable connection between the part OBDM spinless hard-core anyon gas. This allows us to efficiently calculate system with particle numbers much larger than what was previously possible. Given OBDM, we easily momentum distribution system, which again related ayone

The concept of conditional computation for deep nets has been proposed previously to improve model performance by selectively using only parts the conditioned on sample it is processing. In this paper, we investigate input-dependent dynamic filter selection in convolutional neural networks (CNNs). problem interesting because idea forcing different learn from types samples may help us acquire better filters CNNs, generalization and potentially increase interpretability behavior. We propose a...

Abstract Quantum many-body systems in one dimension (1D) exhibit some peculiar properties. In this article, we review of our work on strongly interacting 1D spinor quantum gas. First, discuss a generalized Bose–Fermi mapping that maps the charge degrees freedom to spinless Fermi gas and spin chain model. This also system into weakly one, which is amenable for perturbative calculations. Next, based mapping, construct an ansatz wavefunction system, using many physical quantities can be...

Mismatches between pre-qualified existing components and the particular reuse context in applications have been a major factor hindering component reusability. Although adaptation has acted as key solution of eliminating these mismatches, it often appears impossible for at deep level with tolerable code overheads human work. In this paper, we address above problem by realizing Scenario based Generative Adaptation (SAGA) approach .NET framework. The developed prototype tool (SAGA Bench) helps...

Abstract Electrostatic dust transport on airless planetary bodies, such as the Moon, plays a crucial role in shaping their surface environment and influencing evolution of materials. To investigate effects electron irradiation characteristics micron-sized particles under simulated lunar conditions, we measured diameters velocities moving olivine using laser Doppler system. Experiments were conducted with energies up to 500 eV currents ranging from 1 μ A. The results demonstrate strong...

The opacity, and its Planck Rosseland mean values, of the hot dense Au plasma in local thermodynamics equilibrium are studied by Monte Carlo method based on unresolved transition array (UTA) approximation. average ion model Saha equation used to determine atomic level populations. result gives a more detailed structure for frequency-dependent opacity than popularly super or UTA photon energy range 500 eV 2000 eV. can give better that UTA, with almost same computation effort.

The design process of user interfaces (UIs) often begins with articulating high-level goals. Translating these goals into concrete mock-ups, however, requires extensive effort and UI expertise. To facilitate this for app designers developers, we introduce three deep-learning techniques to create low-fidelity mock-ups from a natural language phrase that describes the goal (e.g. "pop up displaying an image other options"). In particular, contribute two retrieval-based methods one generative...

Non-trivial trails in choosing the right data mining algorithm and tuning its arguments for a set is one of most trivial tasks users. Sharing users' experiences using tools can give valuable clues new user. This paper presents description framework provenance grid platform design two kinds methods representing provenance: WSDL based RDF ones. We also discuss related services those data. The real applications BillionGrid demonstrate effectiveness usefulness.