arXiv subjects

Rui Sun

Publications and source records attributed to Rui Sun.

At least 145 records · Page 8Linked to original sources

arXiv2020

MetaAugment: Sample-Aware Data Augmentation Policy Learning

Automated data augmentation has shown superior performance in image recognition. Existing works search for dataset-level augmentation policies without considering individual sample variations, which are likely to be sub-optimal. On the other hand, learning different policies for different samples naively could greatly increase the computing cost. In this paper, we learn a sample-aware data augmentation policy efficiently by formulating it as a sample reweighting problem. Specifically, an augmentation policy network takes a transformation and the corresponding augmented image as inputs, and outputs a weight to adjust the augmented image loss computed by a task network. At training stage, the task network minimizes the weighted losses of augmented training images, while the policy network minimizes the loss of the task network on a validation set via meta-learning. We theoretically prove the convergence of the training procedure and further derive the exact convergence rate. Superior performance is achieved on widely-used benchmarks including CIFAR-10/100, Omniglot, and ImageNet.

cs.LG↗

arXiv2020

DecAug: Out-of-Distribution Generalization via Decomposed Feature Representation and Semantic Augmentation

While deep learning demonstrates its strong ability to handle independent and identically distributed (IID) data, it often suffers from out-of-distribution (OoD) generalization, where the test data come from another distribution (w.r.t. the training one). Designing a general OoD generalization framework to a wide range of applications is challenging, mainly due to possible correlation shift and diversity shift in the real world. Most of the previous approaches can only solve one specific distribution shift, such as shift across domains or the extrapolation of correlation. To address that, we propose DecAug, a novel decomposed feature representation and semantic augmentation approach for OoD generalization. DecAug disentangles the category-related and context-related features. Category-related features contain causal information of the target object, while context-related features describe the attributes, styles, backgrounds, or scenes, causing distribution shifts between training and test data. The decomposition is achieved by orthogonalizing the two gradients (w.r.t. intermediate features) of losses for predicting category and context labels. Furthermore, we perform gradient-based augmentation on context-related features to improve the robustness of the learned representations. Experimental results show that DecAug outperforms other state-of-the-art methods on various OoD datasets, which is among the very few methods that can deal with different types of OoD generalization challenges.

cs.LG↗

arXiv2020

EPI-based Oriented Relation Networks for Light Field Depth Estimation

Light field cameras record not only the spatial information of observed scenes but also the directions of all incoming light rays. The spatial and angular information implicitly contain geometrical characteristics such as multi-view or epipolar geometry, which can be exploited to improve the performance of depth estimation. An Epipolar Plane Image (EPI), the unique 2D spatial-angular slice of the light field, contains patterns of oriented lines. The slope of these lines is associated with the disparity. Benefiting from this property of EPIs, some representative methods estimate depth maps by analyzing the disparity of each line in EPIs. However, these methods often extract the optimal slope of the lines from EPIs while ignoring the relationship between neighboring pixels, which leads to inaccurate depth map predictions. Based on the observation that an oriented line and its neighboring pixels in an EPI share a similar linear structure, we propose an end-to-end fully convolutional network (FCN) to estimate the depth value of the intersection point on the horizontal and vertical EPIs. Specifically, we present a new feature-extraction module, called Oriented Relation Module (ORM), that constructs the relationship between the line orientations. To facilitate training, we also propose a refocusing-based data augmentation method to obtain different slopes from EPIs of the same scene point. Extensive experiments verify the efficacy of learning relations and show that our approach is competitive to other state-of-the-art methods. The code and the trained models are available at https://github.com/lkyahpu/EPI_ORM.git.

cs.CV↗

arXiv2020

Large spin to charge conversion in topological superconductor \b{eta}-PdBi2 at room temperature

\b{eta}-PdBi2 has attracted much attention for its prospective ability to possess simultaneously topological surface and superconducting states due to its unprecedented spin-orbit interaction (SOC). Whereas most works have focused solely on investigating its topological surface states, the coupling between spin and charge degrees of freedom in this class of quantum material remains unexplored. Here we first report a study of spin-to-charge conversion in a \b{eta}-PdBi2 ultrathin film grown by molecular beam epitaxy, utilizing a spin pumping technique to perform inverse spin Hall effect measurements. We find that the room temperature spin Hall angle of Fe/\b{eta}-PdBi2, θ_SH=0.037. This value is one order of magnitude larger than that of reported conventional superconductors, and is comparable to that of the best SOC metals and topological insulators. Our results provide an avenue for developing superconductor-based spintronic applications.

cond-mat.mes-hall↗

arXiv2020

Revisiting the Supersymmetric Pati-Salam Models from Intersecting D6-branes

Employing new scanning methods, we revisit the systematic construction of three-family $N=1$ supersymmetric Pati-Salam models in Type IIA orientifolds on ${\bf T}^6/(\mathbb Z_2\times \mathbb Z_2)$ with intersecting D6-branes. Arising from the stacks of D6-branes with $U(n)$ gauge symmetries, the Pati-Salam gauge symmetries $SU(4)_C\times SU(2)_L \times SU(2)_R$ can be broken down to the Standard Model via D-brane splitting as well as D- and F-flatness preserving Higgs mechanism. Also, the hidden sector contains $USp(n)$ branes, which are parallel with the orientifold planes or their ${\mathbb Z_2}$ images. We find that the Type II T-duality in the previous study is not an equivalent relation in Pati-Salam model building if the model is not invariant under $SU(2)_L$ and $SU(2)_R$ exchange, and provides a way to obtain new models. We systematically construct the new models with three families, which usually do not have gauge coupling unification at the string scale. We for the first time construct the Pati-Salam models with one wrapping number equal to $5$. In particular, we find that these models carry more refined gauge couplings, and thus with more possibility to have approximate gauge coupling unification.

hep-th↗

arXiv2020

Generalized Supersymmetric Pati-Salam Models from Intersecting D6-branes

Following the scanning methods of arXiv:1910.04530, we for the first time systematically construct the $N=1$ supersymmetric $SU(12)_C\times SU(2)_L\times SU(2)_R$ models, $SU(4)_C\times SU(6)_L\times SU(2)_R$ models, and $SU(4)_C\times SU(2)_L\times SU(6)_R$ models from the Type IIA orientifolds on $\bf T^6/(\mathbb Z_2\times \mathbb Z_2)$ with intersecting D6-branes. These gauge symmetries can be broken down to the Pati-Salam gauge symmetry $SU(4)_C\times SU(2)_L \times SU(2)_R$ via three $SU(12)_C/SU(6)_L/SU(6)_R$ adjoint representation Higgs fields, and further down to the Standard Model (SM) via the D-brane splitting and Higgs mechanism. Also, we obtain three families of the SM fermions, and have the left-handed and right-handed three-family SM fermion unification in the $SU(12)_C\times SU(2)_L\times SU(2)_R$ models, the left-handed three-family SM fermion unification in the $SU(4)_C\times SU(6)_L\times SU(2)_R$ models, and the right-handed three-family SM fermion unification in the $SU(4)_C\times SU(2)_L\times SU(6)_R$ models. Moreover, the $SU(4)_C\times SU(6)_L\times SU(2)_R$ models and $SU(4)_C\times SU(2)_L\times SU(6)_R$ models are related by the left and right gauge symmetry exchanging, as well as a variation of type II T-duality. Furthermore, the hidden sector contains $USp(n)$ branes, which are parallel with the orientifold planes or their $Z_2$ images and might break the supersymmetry via gaugino condensations.

hep-th↗

arXiv2020

Optimal Path Planning for Automated Dimensional Inspection of Free-Form Surfaces

Structural dimensional inspection is vital for the process monitoring, quality control, and fault diagnosis in the mass production of auto bodies. Comparing with the non-contact measurement, the high-precision five-axis measuring machine with the touch-trigger probe is a preferred choice for data collection. It can assist manufacturers in making accurate inspection quickly. As the increase of free-form surfaces and diverse surface orientations in auto body design, existing inspection approaches cannot capture some new critical features in the curvature of auto bodies in an efficient way. Therefore, we need to develop new path planning methods for automated dimensional inspection of free-form surfaces. This paper proposes an optimal path planning system for automated programming of measuring point inspection by incorporating probe rotations and effective collision detection. Specifically, the methodological contributions include: (i) a dynamic searching volume-based algorithm is developed to detect potential collisions in the local path between measurement points; (ii) a local path generation method is proposed with the integration of the probe trajectory and the stylus rotation. Then, the inspection time matrix is proposed to quantify the measuring time of diverse local paths; (iii) an optimization approach of the global inspection path for all critical points on the auto body is developed to minimize the total inspection time. A case study has been conducted on an auto body to verify the performance of the proposed method. Results show that the collision-free path for the free-form auto body could be generated automatically with off-line programming, and the proposed method produces about 40% fewer dummy points and needs 32% less movement time in the auto body inspection process.

cs.RO↗

arXiv2020

Online Learning and Optimization for Revenue Management Problems with Add-on Discounts

We study in this paper a revenue management problem with add-on discounts. The problem is motivated by the practice in the video game industry, where a retailer offers discounts on selected supportive products (e.g. video games) to customers who have also purchased the core products (e.g. video game consoles). We formulate this problem as an optimization problem to determine the prices of different products and the selection of products with add-on discounts. To overcome the computational challenge of this optimization problem, we propose an efficient FPTAS algorithm that can solve the problem approximately to any desired accuracy. Moreover, we consider the revenue management problem in the setting where the retailer has no prior knowledge of the demand functions of different products. To resolve this problem, we propose a UCB-based learning algorithm that uses the FPTAS optimization algorithm as a subroutine. We show that our learning algorithm can converge to the optimal algorithm that has access to the true demand functions, and we prove that the convergence rate is tight up to a certain logarithmic term. In addition, we conduct numerical experiments with the real-world transaction data we collect from a popular video gaming brand's online store on Tmall.com. The experiment results illustrate our learning algorithm's robust performance and fast convergence in various scenarios. We also compare our algorithm with the optimal policy that does not use any add-on discount, and the results show the advantages of using the add-on discount strategy in practice.

cs.DS↗

arXiv2020

Real Entropy Can Also Predict Daily Voice Traffic for Wireless Network Users

Voice traffic prediction is significant for network deployment optimization thus to improve the network efficiency. The real entropy based theorectical bound and corresponding prediction models have demonstrated their success in mobility prediction. In this paper, the real entropy based predictability analysis and prediction models are introduced into voice traffic prediction. For this adoption, the traffic quantification methods is proposed and discussed. Based on the real world voice traffic data, the prediction accuracy of N-order Markov models, diffusion based model and MF model are presented, among which, 25-order Markov models performs best and approach close to the maximum predictability. This work demonstrates that, the real entropy can also predict voice traffic well which broaden the understanding on the real entropy based prediction theory.

cs.NI↗

arXiv2020

Influence of Laser Intensity Fluctuation on Single-Cesium Atom Trapping Lifetime in a 1064-nm Microscopic Optical Tweezer

An optical tweezer composed of a strongly focused single-spatial-mode Gaussian beam of a red-detuned 1064-nm laser can confine a single-cesium (Cs) atom at the strongest point of the light intensity. We can use this for coherent manipulation of single-quantum bits and single-photon sources. The trapping lifetime of the atoms in the optical tweezers is very short due to the impact of the background atoms, the laser intensity fluctuation of optical tweezer and the residual thermal motion of the atoms. In this paper, we analyzed the influence of the background pressure, the trap frequency of optical tweezers and the parametric heating of the optical tweezer on the atomic trapping lifetime. Combined with the external feedback loop based on an acousto-optical modulator (AOM), the intensity fluctuation of the 1064-nm laser in the time domain was suppressed from $\pm$ 3.360$\%$ to $\pm$ 0.064$\%$, and the suppression bandwidth reached approximately 33 kHz. The trapping lifetime of a single Cs atom in the microscopic optical tweezer was extended from 4.04 s to 6.34 s.

physics.atom-ph↗

arXiv2019

M5 branes and Theta Functions

We propose quantum states for Little String Theories (LSTs) arising from M5 branes probing A- and D-type singularities. This extends Witten's picture of M5 brane partition functions as theta functions to this more general setup. Compactifying the world-volume of the five-branes on a two-torus, we find that the corresponding theta functions are sections of line bundles over complex 4-tori. This formalism allows us to derive Seiberg-Witten curves for the resulting four-dimensional theories. Along the way, we prove a duality for LSTs observed by Iqbal, Hohenegger and Rey.

hep-th↗

arXiv2019

On Missing Bianchi Identities in Cohomology Formulation

In this article, we perform a deep analysis of the Bianchi identities in the two known formulations developed for the four-dimensional effective type IIA supergravity theory with (non-)geometric fluxes. In what we call the `first formulation', fluxes are expressed in the real six-dimensional indices while in the `second formulation', fluxes are written in the cohomology form. We find that the set of flux constraints arising from these two known formulations are not equivalent, and there are missing identities in the cohomology version which need to be supplemented to match with the first formulation. By analyzing two explicit examples, we conjecture a model independent form for (the most of) the missing identities. These identities have been mostly overlooked in the previous attempts of studying moduli stabilization, particularly for the models developed in the beyond toroidal frameworks, where they could play some important role.

hep-th↗

arXiv2019

RANS Equations with Explicit Data-Driven Reynolds Stress Closure Can Be Ill-Conditioned

Reynolds-averaged Navier--Stokes (RANS) simulations with turbulence closure models continue to play important roles in industrial flow simulations. However, the commonly used linear eddy viscosity models are intrinsically unable to handle flows with non-equilibrium turbulence. Reynolds stress models, on the other hand, are plagued by their lack of robustness. Recent studies in plane channel flows found that even substituting Reynolds stresses with errors below 0.5% from direct numerical simulation (DNS) databases into RANS equations leads to velocities with large errors (up to 35%). While such an observation may have only marginal relevance to traditional Reynolds stress models, it is disturbing for the recently emerging data-driven models that treat the Reynolds stress as an explicit source term in the RANS equations, as it suggests that the RANS equations with such models can be ill-conditioned. So far, a rigorous analysis of the condition of such models is still lacking. As such, in this work we propose a metric based on local condition number function for a priori evaluation of the conditioning of the RANS equations. We further show that the ill-conditioning cannot be explained by the global matrix condition number of the discretized RANS equations. Comprehensive numerical tests are performed on turbulent channel flows at various Reynolds numbers and additionally on two complex flows, i.e., flow over periodic hills and flow in a square duct. Results suggest that the proposed metric can adequately explain observations in previous studies, i.e., deteriorated model conditioning with increasing Reynolds number and better conditioning of the implicit treatment of Reynolds stress compared to the explicit treatment. This metric can play critical roles in the future development of data-driven turbulence models by enforcing the conditioning as a requirement on these models.

physics.flu-dyn↗

arXiv2018

Multi-view Point Cloud Registration with Adaptive Convergence Threshold and its Application on 3D Model Retrieval

Multi-view point cloud registration is a hot topic in the communities of multimedia technology and artificial intelligence (AI). In this paper, we propose a framework to reconstruct the 3D models by the multi-view point cloud registration algorithm with adaptive convergence threshold, and subsequently apply it to 3D model retrieval. The iterative closest point (ICP) algorithm is implemented combining with the motion average algorithm for the registration of multi-view point clouds. After the registration process, we design applications for 3D model retrieval. The geometric saliency map is computed based on the vertex curvature. The test facial triangle is then generated based on the saliency map, which is applied to compare with the standard facial triangle. The face and non-face models are then discriminated. The experiments and comparisons prove the effectiveness of the proposed framework.

cs.CV↗

arXiv2018

S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction

Recent studies have used deep residual convolutional neural networks (CNNs) for JPEG compression artifact reduction. This study proposes a scalable CNN called S-Net. Our approach effectively adjusts the network scale dynamically in a multitask system for real-time operation with little performance loss. It offers a simple and direct technique to evaluate the performance gains obtained with increasing network depth, and it is helpful for removing redundant network layers to maximize the network efficiency. We implement our architecture using the Keras framework with the TensorFlow backend on an NVIDIA K80 GPU server. We train our models on the DIV2K dataset and evaluate their performance on public benchmark datasets. To validate the generality and universality of the proposed method, we created and utilized a new dataset, called WIN143, for over-processed images evaluation. Experimental results indicate that our proposed approach outperforms other CNN-based methods and achieves state-of-the-art performance.

cs.CV↗

arXiv2018

Symplectic formulation of the type IIA nongeometric scalar potential

We study the four-dimensional (4D) scalar potential arising from a generalized type IIA flux superpotential including the (non-)geometric fluxes. First, we show that using a set of peculiar flux combinations, the 4D scalar potential can be formulated into a very compact form. This is what we call as the `symplectic formulation' from which one could easily anticipate the ten-dimensional origin of the effective scalar potential. We support our formulation through an alternate derivation of the scalar potential via considering the Double Field Theory (DFT) reduction on a generic Calabi Yau orientifold. In addition, we also exemplify the insights of our formulation with explicit computations for two concrete toroidal examples using orientifolds of the complex threefolds ${\mathbb T}^6/{({\mathbb Z}_2 \times {\mathbb Z}_2)}$ and ${\mathbb T}^6/{\mathbb Z}_4$.

hep-th↗

arXiv2018

Particle dynamics in self-generated dunes over a range of hydraulic and sediment transport conditions using LES--DEM

Direct measurement of vertical and longitudinal sediment fluxes on migrating sandy bedforms are extremely difficult to perform in both the field and laboratory. In this study we use the LES--DEM (large eddy simulation--discrete element method) solver SediFoam to examine the individual particle motions and resulting fluxes in a domain of self-generated dunes. In SediFoam, the motions of, and collisions among, the sediment grains as well as their interactions with surrounding turbulent flows are resolved. The numerical simulations are performed over a range of transport settings, spanning bedform inception through washout conditions, to examine the individual particle dynamics. The space-time evolution of dune surfaces is demonstrated. The self-generated dunes are stable at relatively low Reynolds numbers, but then become increasingly unstable at higher Reynolds numbers; eventually washing out as the number of bypass grains and particles in suspension increase. Data from the simulation are used to examine the vertical entrainment rate of particles and the fractionation of total sediment load into bed and suspended fractions as a function of transport conditions. The study also compares the sediment transport rate obtained using the bedform geometry and celerity to the true transport rate at different transport stages.

physics.geo-ph↗

arXiv2017

Study of sedimentation of non-cohesive particles via CFD-DEM simulations

The sedimentation process of granular materials exists ubiquitously in nature and many fields which involve the solid-liquid separation. This paper employs the coupled computational fluid dynamics and discrete element method (CFD-DEM) to investigate the sedimentation process of non-cohesive particles, including the hindered settling stage and the deposition stage. Firstly, the coupled CFD-DEM model for sedimentation is validated by the hindered settling velocity at different solid volume concentrations of suspension (0.05~0.6). Two typical modes of sedimentation are also presented by the concentration profiles and the equal-concentration lines. Then, the comparisons between mono- and poly-dispersed particle system are detailed. In the sedimentation of the poly-dispersed particle system, the segregation phenomenon is simulated. Furthermore, this segregation effect reduces with the increase of the initial solid concentration of suspension. From the simulations, the contact force between every pair of particles can be obtained, hence we demonstrate the "effective stress principle" from the view of the particle contact force by giving the correspondence between the particle contact force and the "effective stress", which is a critical concept of soil mechanics. We also demonstrate the effective stress principle from the view of the contact force acting on particles. Moreover, the deposition stage can be simulated by CFD-DEM method, therefore the solid concentrations of sediment bed on different conditions are studied. Based on the simulation results of maximum solid concentration and the theory of sedimentation, this paper also discusses a method to calculate the critical time when sedimentation ends of two typical modes of sedimentation.

physics.flu-dyn↗