arXiv subjects

Jing Yang

Publications and source records attributed to Jing Yang.

At least 271 records · Page 15Linked to original sources

arXiv2023

A Mountain-Shaped Single-Stage Network for Accurate Image Restoration

Image restoration is the task of aiming to obtain a high-quality image from a corrupt input image, such as deblurring and deraining. In image restoration, it is typically necessary to maintain a complex balance between spatial details and contextual information. Although a multi-stage network can optimally balance these competing goals and achieve significant performance, this also increases the system's complexity. In this paper, we propose a mountain-shaped single-stage design base on a simple U-Net architecture, which removes or replaces unnecessary nonlinear activation functions to achieve the above balance with low system complexity. Specifically, we propose a feature fusion middleware (FFM) mechanism as an information exchange component between the encoder-decoder architectural levels. It seamlessly integrates upper-layer information into the adjacent lower layer, sequentially down to the lowest layer. Finally, all information is fused into the original image resolution manipulation level. This preserves spatial details and integrates contextual information, ensuring high-quality image restoration. In addition, we propose a multi-head attention middle block (MHAMB) as a bridge between the encoder and decoder to capture more global information and surpass the limitations of the receptive field of CNNs. Extensive experiments demonstrate that our approach, named as M3SNet, outperforms previous state-of-the-art models while using less than half the computational costs, for several image restoration tasks, such as image deraining and deblurring.

cs.CV↗

arXiv2023

Bézout Subresultants for Univariate Polynomials in General Basis

Subresultant is a powerful tool for developing various algorithms in computer algebra. Subresultants for polynomials in standard basis (i.e., power basis) have been well studied so far. With the popularity of basis-preserving algorithms, resultants and subresultants in non-standard basis are drawing more and more attention. In this paper, we develop a formula for Bézout subresultants of univariate polynomials in general basis, which covers a broad range of non-standard bases. More explicitly, the input polynomials are provided in a given general basis and the resulting subresultants are Bézout-type expressions in the same basis. It is shown that the subresultants share the essential properties as the subresultants in standard basis.

cs.SC↗

arXiv2023

Subresultant of several univariate polynomials

Subresultant of two univariate polynomials is a fundamental object in computational algebra and geometry with many applications (for instance, parametric GCD and parametric multiplicity of roots). In this paper, we generalize the theory of subresultants of two polynomials to arbitrary number of polynomials, resulting in multi-polynomial subresultants. Specifically, 1. we propose a definition of multi-polynomial subresultants, which is an expression in terms of roots; 2. we illustrate the usefulness of the proposed definition via the following two fundamental applications: - parametric GCD of multi-polynomials, and - parametric multiplicity of roots of a polynomial; 3. we provide several expressions for the multi-polynomials subresultants in terms of coefficients, for computation.

cs.SC↗

arXiv2023

Parametric "Non-nested" Discriminants for Multiplicities of Univariate Polynomials

We consider the problem of complex root classification, i.e., finding the conditions on the coefficients of a univariate polynomial for all possible multiplicity structures on its complex roots. It is well known that such conditions can be written as conjunctions of several polynomial equations and one inequation in the coefficients. Those polynomials in the coefficients are called discriminants for multiplicities. It is well known that discriminants can be obtained by using repeated parametric gcd's. The resulting discriminants are usually nested determinants, that is, determinants of matrices whose entries are determinants, and so son. In this paper, we give a new type of discriminants which are not based on repeated gcd's. The new discriminants are simpler in that they are non-nested determinants and have smaller maximum degrees.

cs.SC↗

arXiv2023

Unicom: Universal and Compact Representation Learning for Image Retrieval

Modern image retrieval methods typically rely on fine-tuning pre-trained encoders to extract image-level descriptors. However, the most widely used models are pre-trained on ImageNet-1K with limited classes. The pre-trained feature representation is therefore not universal enough to generalize well to the diverse open-world classes. In this paper, we first cluster the large-scale LAION400M into one million pseudo classes based on the joint textual and visual features extracted by the CLIP model. Due to the confusion of label granularity, the automatically clustered dataset inevitably contains heavy inter-class conflict. To alleviate such conflict, we randomly select partial inter-class prototypes to construct the margin-based softmax loss. To further enhance the low-dimensional feature representation, we randomly select partial feature dimensions when calculating the similarities between embeddings and class-wise prototypes. The dual random partial selections are with respect to the class dimension and the feature dimension of the prototype matrix, making the classification conflict-robust and the feature embedding compact. Our method significantly outperforms state-of-the-art unsupervised and supervised image retrieval approaches on multiple benchmarks. The code and pre-trained models are released to facilitate future research https://github.com/deepglint/unicom.

cs.CV↗

arXiv2023

Light Sampling Field and BRDF Representation for Physically-based Neural Rendering

Physically-based rendering (PBR) is key for immersive rendering effects used widely in the industry to showcase detailed realistic scenes from computer graphics assets. A well-known caveat is that producing the same is computationally heavy and relies on complex capture devices. Inspired by the success in quality and efficiency of recent volumetric neural rendering, we want to develop a physically-based neural shader to eliminate device dependency and significantly boost performance. However, no existing lighting and material models in the current neural rendering approaches can accurately represent the comprehensive lighting models and BRDFs properties required by the PBR process. Thus, this paper proposes a novel lighting representation that models direct and indirect light locally through a light sampling strategy in a learned light sampling field. We also propose BRDF models to separately represent surface/subsurface scattering details to enable complex objects such as translucent material (i.e., skin, jade). We then implement our proposed representations with an end-to-end physically-based neural face skin shader, which takes a standard face asset (i.e., geometry, albedo map, and normal map) and an HDRI for illumination as inputs and generates a photo-realistic rendering as output. Extensive experiments showcase the quality and efficiency of our PBR face skin shader, indicating the effectiveness of our proposed lighting and material representations.

cs.CV↗

arXiv2023

Intersections of linear codes and related MDS codes with new Galois hulls

Let $\mathrm{SLAut}(\mathbb{F}_{q}^{n})$ denote the group of all semilinear isometries on $\mathbb{F}_{q}^{n}$, where $q=p^{e}$ is a prime power. In this paper, we investigate general properties of linear codes associated with $σ$ duals for $σ\in\mathrm{SLAut}(\mathbb{F}_{q}^{n})$. We show that the dimension of the intersection of two linear codes can be determined by generator matrices of such codes and their $σ$ duals. We also show that the dimension of $σ$ hull of a linear code can be determined by a generator matrix of it or its $σ$ dual. We give a characterization on $σ$ dual and $σ$ hull of a matrix-product code. We also investigate the intersection of a pair of matrix-product codes. We provide a necessary and sufficient condition under which any codeword of a generalized Reed-Solomon (GRS) code or an extended GRS code is contained in its $σ$ dual. As an application, we construct eleven families of $q$-ary MDS codes with new $\ell$-Galois hulls satisfying $2(e-\ell)\mid e$, which are not covered by the latest papers by Cao (IEEE Trans. Inf. Theory 67(12), 7964-7984, 2021) and by Fang et al. (Cryptogr. Commun. 14(1), 145-159, 2022) when $\ell\neq \frac{e}{2}$.

cs.IT↗

arXiv2023

On two conserved quantities in the inviscid electron and Hall magnetohydrodynamic equations

In this paper, we are concerned with the energy and magnetic helicity conservation of weak solutions for both the electron and Hall magnetohydrodynamic equations. Various sufficient criteria to ensure the energy and magnetic helicity conservation in Onsager's critical spaces $\underline{B}^α_{p,VMO}$ and $B^α_{p,c(\mathbb{N})}$ in these systems are established. Moreover, for the E-MHD equations, we observe that the conservation criteria of energy and magnetic helicity to the E-MHD equations correspond to the helicity and energy to the ideal incompressible Euler equations, respectively.

math.AP↗

arXiv2023

Safe Exploration Incurs Nearly No Additional Sample Complexity for Reward-free RL

Reward-free reinforcement learning (RF-RL), a recently introduced RL paradigm, relies on random action-taking to explore the unknown environment without any reward feedback information. While the primary goal of the exploration phase in RF-RL is to reduce the uncertainty in the estimated model with minimum number of trajectories, in practice, the agent often needs to abide by certain safety constraint at the same time. It remains unclear how such safe exploration requirement would affect the corresponding sample complexity in order to achieve the desired optimality of the obtained policy in planning. In this work, we make a first attempt to answer this question. In particular, we consider the scenario where a safe baseline policy is known beforehand, and propose a unified Safe reWard-frEe ExploraTion (SWEET) framework. We then particularize the SWEET framework to the tabular and the low-rank MDP settings, and develop algorithms coined Tabular-SWEET and Low-rank-SWEET, respectively. Both algorithms leverage the concavity and continuity of the newly introduced truncated value functions, and are guaranteed to achieve zero constraint violation during exploration with high probability. Furthermore, both algorithms can provably find a near-optimal policy subject to any constraint in the planning phase. Remarkably, the sample complexities under both algorithms match or even outperform the state of the art in their constraint-free counterparts up to some constant factors, proving that safety constraint hardly increases the sample complexity for RF-RL.

cs.LG↗

arXiv2023

Improved Sample Complexity for Reward-free Reinforcement Learning under Low-rank MDPs

In reward-free reinforcement learning (RL), an agent explores the environment first without any reward information, in order to achieve certain learning goals afterwards for any given reward. In this paper we focus on reward-free RL under low-rank MDP models, in which both the representation and linear weight vectors are unknown. Although various algorithms have been proposed for reward-free low-rank MDPs, the corresponding sample complexity is still far from being satisfactory. In this work, we first provide the first known sample complexity lower bound that holds for any algorithm under low-rank MDPs. This lower bound implies it is strictly harder to find a near-optimal policy under low-rank MDPs than under linear MDPs. We then propose a novel model-based algorithm, coined RAFFLE, and show it can both find an $ε$-optimal policy and achieve an $ε$-accurate system identification via reward-free exploration, with a sample complexity significantly improving the previous results. Such a sample complexity matches our lower bound in the dependence on $ε$, as well as on $K$ in the large $d$ regime, where $d$ and $K$ respectively denote the representation dimension and action space cardinality. Finally, we provide a planning algorithm (without further interaction with true environment) for RAFFLE to learn a near-accurate representation, which is the first known representation learning guarantee under the same setting.

cs.LG↗

arXiv2023

Dynamic Multi-View Fusion Mechanism For Chinese Relation Extraction

Recently, many studies incorporate external knowledge into character-level feature based models to improve the performance of Chinese relation extraction. However, these methods tend to ignore the internal information of the Chinese character and cannot filter out the noisy information of external knowledge. To address these issues, we propose a mixture-of-view-experts framework (MoVE) to dynamically learn multi-view features for Chinese relation extraction. With both the internal and external knowledge of Chinese characters, our framework can better capture the semantic information of Chinese characters. To demonstrate the effectiveness of the proposed framework, we conduct extensive experiments on three real-world datasets in distinct domains. Experimental results show consistent and significant superiority and robustness of our proposed framework. Our code and dataset will be released at: https://gitee.com/tmg-nudt/multi-view-of-expert-for-chineserelation-extraction

cs.CL↗

arXiv2023

Proton-fluence dependent magnetic properties of exfoliable quasi-2D van der Waals Cr2Si2Te6 magnet

The discovery of long-range magnetic ordering in atomically thin materials catapulted the van der Waals (vdW) family of compounds into an unprecedented popularity. In particular, with a current push in space exploration, it is beneficial to study how the properties of such materials evolve under proton irradiation. Owing to their robust intra-layer stability and sensitivity to external perturbations, these materials provide excellent opportunities for studying proton irradiation as a non-destructive tool for controlling their magnetic properties. Specifically, the exfoliable Cr2Si2Te6 (CST) is a ferromagnetic semiconductor with the Curie temperature (TC) of ~32 K. Here, we have investigated the magnetic properties of CST upon proton irradiation as a function of fluence (1 x 1015, 5 x 1015, 1 x 1016, 5 x 1016, and 1 x 1018 H+/cm2) by employing variable-temperature, variable-field magnetization measurements coupled with electron paramagnetic resonance (EPR) spectroscopy and detail how the magnetization, magnetic anisotropy and EPR spectral parameters vary as a function of proton fluence across the magnetic phase transition. While the TC remains constant as a function of proton fluence, we observed that the saturation magnetization and magnetic anisotropy diverge at the proton fluence of 5 x 1016 H+/cm2, which is prominent in the ferromagnetic phase, in particular. This work demonstrates that proton irradiation is a feasible method for modifying the magnetic properties and local magnetic interactions of vdWs crystals, which represents a significant step forward in the design of future spintronic and magneto-electronic applications.

cond-mat.mtrl-sci↗

arXiv2023

ConMAE: Contour Guided MAE for Unsupervised Vehicle Re-Identification

Vehicle re-identification is a cross-view search task by matching the same target vehicle from different perspectives. It serves an important role in road-vehicle collaboration and intelligent road control. With the large-scale and dynamic road environment, the paradigm of supervised vehicle re-identification shows limited scalability because of the heavy reliance on large-scale annotated datasets. Therefore, the unsupervised vehicle re-identification with stronger cross-scene generalization ability has attracted more attention. Considering that Masked Autoencoder (MAE) has shown excellent performance in self-supervised learning, this work designs a Contour Guided Masked Autoencoder for Unsupervised Vehicle Re-Identification (ConMAE), which is inspired by extracting the informative contour clue to highlight the key regions for cross-view correlation. ConMAE is implemented by preserving the image blocks with contour pixels and randomly masking the blocks with smooth textures. In addition, to improve the quality of pseudo labels of vehicles for unsupervised re-identification, we design a label softening strategy and adaptively update the label with the increase of training steps. We carry out experiments on VeRi-776 and VehicleID datasets, and a significant performance improvement is obtained by the comparison with the state-of-the-art unsupervised vehicle re-identification methods. The code is available on the website of https://github.com/2020132075/ConMAE.

cs.CV↗

arXiv2023

Spin-orbit-coupled spinor gap solitons in Bose-Einstein condensates

Spin-1 spin-orbit-coupled spinor Bose-Einstein condensates have been realized in experiment. We study spin-orbit-coupled spinor gap solitons in this experimentally realizable system with an optical lattice. The spin-dependent parity symmetry of the spin-orbit coupling plays an important role in properties of gap solitons. Two families of solitons with opposite spin-dependent parity are found. Using an approximating model by replacing the optical lattice with a Harmonic trap, we demonstrate the physical origination of these two families. For a large negative quadratic Zeeman shift, we also find a type of gap solitons that spontaneously breaks the spin-dependent parity symmetry due to the ferromagnetic spinor interactions.

cond-mat.quant-gas↗

arXiv2023

Adaptive Least-Squares Methods for Convection-Dominated Diffusion-Reaction Problems

This paper studies adaptive least-squares finite element methods for convection-dominated diffusion-reaction problems. The least-squares methods are based on the first-order system of the primal and dual variables with various ways of imposing outflow boundary conditions. The coercivity of the homogeneous least-squares functionals are established, and the a priori error estimates of the least-squares methods are obtained in a norm that incorporates the streamline derivative. All methods have the same convergence rate provided that meshes in the layer regions are fine enough. To increase computational accuracy and reduce computational cost, adaptive least-squares methods are implemented and numerical results are presented for some test problems.

math.NA↗

arXiv2023

Determinate Node Selection for Semi-supervised Classification Oriented Graph Convolutional Networks

Graph Convolutional Networks (GCNs) have been proved successful in the field of semi-supervised node classification by extracting structural information from graph data. However, the random selection of labeled nodes used by GCNs may lead to unstable generalization performance of GCNs. In this paper, we propose an efficient method for the deterministic selection of labeled nodes: the Determinate Node Selection (DNS) algorithm. The DNS algorithm identifies two categories of representative nodes in the graph: typical nodes and divergent nodes. These labeled nodes are selected by exploring the structure of the graph and determining the ability of the nodes to represent the distribution of data within the graph. The DNS algorithm can be applied quite simply on a wide range of semi-supervised graph neural network models for node classification tasks. Through extensive experimentation, we have demonstrated that the incorporation of the DNS algorithm leads to a remarkable improvement in the average accuracy of the model and a significant decrease in the standard deviation, as compared to the original method.

cs.LG↗

arXiv2022

Transformer-based Cross-Modal Recipe Embeddings with Large Batch Training

In this paper, we present a cross-modal recipe retrieval framework, Transformer-based Network for Large Batch Training (TNLBT), which is inspired by ACME~(Adversarial Cross-Modal Embedding) and H-T~(Hierarchical Transformer). TNLBT aims to accomplish retrieval tasks while generating images from recipe embeddings. We apply the Hierarchical Transformer-based recipe text encoder, the Vision Transformer~(ViT)-based recipe image encoder, and an adversarial network architecture to enable better cross-modal embedding learning for recipe texts and images. In addition, we use self-supervised learning to exploit the rich information in the recipe texts having no corresponding images. Since contrastive learning could benefit from a larger batch size according to the recent literature on self-supervised learning, we adopt a large batch size during training and have validated its effectiveness. In the experiments, the proposed framework significantly outperformed the current state-of-the-art frameworks in both cross-modal recipe retrieval and image generation tasks on the benchmark Recipe1M. This is the first work which confirmed the effectiveness of large batch training on cross-modal recipe embeddings.

cs.CV↗

arXiv2022

Latest data constraint of some parameterized dark energy models

Using various latest cosmological datasets including Type-Ia supernovae, cosmic microwave background radiation, baryon acoustic oscillations, and estimations of the Hubble parameter, we test some dark energy models with parameterized equations of state and try to distinguish or select observation-preferred models. We obtain the best fitting results of the six models and calculate their values of the Akaike Information Criteria and Bayes Information Criterion. And we can distinguish these dark energy models from each other by using these two information criterions. However, the $Λ$CDM model remains the best fit model. Furthermore, we perform geometric diagnostics including statefinder and Om diagnostics to understand the geometric behaviour of the dark energy models. We find that the six DE models can be distinguished from each other and from $Λ$CDM, Chaplygin gas, quintessence models after the statefinder and Om diagnostics were performed. Finally, we consider the growth factor of the dark energy models with comparison to $Λ$CDM model. Still, we find the models can be distinguished from each other and from $Λ$CDM model through the growth factor approximation.

astro-ph.CO↗