Searcharxiv⌕ Search

arXiv subjects

Xiang Chen

Publications and source records attributed to Xiang Chen.

At least 397 records · Page 22Linked to original sources
arXiv2023

FedHC: A Scalable Federated Learning Framework for Heterogeneous and Resource-Constrained Clients

Federated Learning (FL) is a distributed learning paradigm that empowers edge devices to collaboratively learn a global model leveraging local data. Simulating FL on GPU is essential to expedite FL algorithm prototyping and evaluations. However, current FL frameworks overlook the disparity between algorithm simulation and real-world deployment, which arises from heterogeneous computing capabilities and imbalanced workloads, thus misleading evaluations of new algorithms. Additionally, they lack flexibility and scalability to accommodate resource-constrained clients. In this paper, we present FedHC, a scalable federated learning framework for heterogeneous and resource-constrained clients. FedHC realizes system heterogeneity by allocating a dedicated and constrained GPU resource budget to each client, and also simulates workload heterogeneity in terms of framework-provided runtime. Furthermore, we enhance GPU resource utilization for scalable clients by introducing a dynamic client scheduler, process manager, and resource-sharing mechanism. Our experiments demonstrate that FedHC has the capability to capture the influence of various factors on client execution time. Moreover, despite resource constraints for each client, FedHC achieves state-of-the-art efficiency compared to existing frameworks without limits. When subjecting existing frameworks to the same resource constraints, FedHC achieves a 2.75x speedup. Code has been released on https://github.com/if-lab-repository/FedHC.

cs.DC↗
arXiv2023

Probing the neutron-skin of unstable nuclei with heavy ion collisions

To improve the constraints of symmetry energy at subsaturation density, measuring and accumulating more neutron skin data for neutron rich unstable nuclei is naturally required. Aiming to probe the neutron skin of unstable nuclei by using low-intermediate energy heavy ion collisions, we develop a new version of improved quantum molecular dynamics model, in which the neutron skin of the initial nucleus and the mean field potential in nucleon propagation are consistently treated. Our calculations show that the three observables, such as the cross sections of the primary projectile-like residues with $A>100$ ($σ_{A>100}$), the difference of $σ_{A>100}$ between $^{132}$Sn+$^{124}$Sn and $^{124}$Sn+$^{124}$Sn systems ($δσ_{A>100}$), and the neutron to proton yield ratio ($R(n/p)$) in the transverse direction, could be used to measure the neutron skin of the unstable nuclei and to constrain the slope of the symmetry energy in the future.

nucl-th↗
arXiv2023

GACER: Granularity-Aware ConcurrEncy Regulation for Multi-Tenant Deep Learning

As deep learning continues to advance and is applied to increasingly complex scenarios, the demand for concurrent deployment of multiple neural network models has arisen. This demand, commonly referred to as multi-tenant computing, is becoming more and more important. However, even the most mature GPU-based computing systems struggle to adequately address the significant heterogeneity and complexity among concurrent models in terms of resource allocation and runtime scheduling. And this usually results in considerable resource utilization and throughput issues. To tackle these issues, this work proposes a set of optimization techniques that advance the granularity of computing management from both the spatial and temporal perspectives, specifically tailored to heterogeneous model compositions for deep learning inference and training. These techniques are further integrated as GACER -- an automated optimization framework that provides high-utilization, high-throughput, and low-latency multi-tenant computing support. And our experiments demonstrate that GACER significantly improves the overall resource utilization and consistently achieves outstanding speedups compared to native GPU computing frameworks and existing state-of-the-art optimization works.

cs.DC↗
arXiv2023

Probing the nucleon effective mass splitting with the light particle emission

In this work, we first analyze the correlations among different nuclear matter parameters which are obtained by the 119 effective Skyrme interaction sets. The values of the correlation coefficients illustrate that the magnitude of the effective mass splitting is crucial for the tight constraints of the symmetry energy via HICs. Thus, the main object of this work is to investigate the impacts of the effective mass splitting on the heavy ion collisions observables. The $^{86}$Kr+$^{208}$Pb system at the beam energy ranging from 25A MeV to 200A MeV are simulated within the framework of the improved quantum molecular dynamics model (ImQMD-Sky) for this goal. Our calculations show that the slopes of the spectra of $\ln$[Y(n)/Y(p)] and $\ln$[Y(t)/Y($^3$He)], which are the logarithms of the neutron to proton yield ratios and triton to Helium-3 yield ratios, are directly related to the effective mass splitting and can be used to probe the effective mass splitting.

nucl-th↗
arXiv2023

Loop Closure Detection Based on Object-level Spatial Layout and Semantic Consistency

Visual simultaneous localization and mapping (SLAM) systems face challenges in detecting loop closure under the circumstance of large viewpoint changes. In this paper, we present an object-based loop closure detection method based on the spatial layout and semanic consistency of the 3D scene graph. Firstly, we propose an object-level data association approach based on the semantic information from semantic labels, intersection over union (IoU), object color, and object embedding. Subsequently, multi-view bundle adjustment with the associated objects is utilized to jointly optimize the poses of objects and cameras. We represent the refined objects as a 3D spatial graph with semantics and topology. Then, we propose a graph matching approach to select correspondence objects based on the structure layout and semantic property similarity of vertices' neighbors. Finally, we jointly optimize camera trajectories and object poses in an object-level pose graph optimization, which results in a globally consistent map. Experimental results demonstrate that our proposed data association approach can construct more accurate 3D semantic maps, and our loop closure method is more robust than point-based and object-based methods in circumstances with large viewpoint changes.

cs.CV↗
arXiv2023

Role of electron-phonon coupling in excitonic insulator candidate Ta2NiSe5

Electron-hole bound pairs, or excitons, are common excitations in semiconductors. They can spontaneously form and ``condense'' into a new insulating ground state -- the so-called excitonic insulator -- when the energy of electron-hole Coulomb attraction exceeds the band gap. In the presence of electron-phonon coupling, a periodic lattice distortion often concomitantly occurs with this exciton condensation. However, similar structural transition can also be induced by electron-phonon coupling itself, therefore hindering the clean identification of bulk excitonic insulators based on reductionistic reasoning (e.g. which instability is the ``driving force'' of the phase transition). Using high-resolution synchrotron x-ray diffraction and angle-resolved photoemission spectroscopy techniques, we identify key electron-phonon coupling effects in a leading excitonic insulator candidate Ta2NiSe5. These include an extensive unidirectional lattice fluctuation and an electronic pseudogap in the normal state, as well as a negative electronic compressibility in the charge-doped broken-symmetry state. In combination with first principles and model calculations, we determine a minimal lattice model and the corresponding interaction parameters that capture the experimental observations. More importantly, we show how the Coulomb and electron-phonon coupling effects can be separated on the level of lattice model, and demonstrate a general framework beyond the reductionist approach in the investigation of correlated systems with intertwined orders.

cond-mat.str-el↗
arXiv2023

Robust Tracking Control for Nonlinear Systems: Performance optimization via extremum seeking

This paper presents a controller design and optimization framework for nonlinear dynamic systems to track a given reference signal in the presence of disturbances when the task is repeated over a finite-time interval. This novel framework mainly consists of two steps. The first step is to design a robust linear quadratic tracking controller based on the existing control structure with a Youla-type filter $\tilde Q$. Secondly, an extra degree of freedom: a parameterization in terms of $\tilde Q$, is added to this design framework. This extra design parameter is tuned iteratively from measured tracking cost function with the given disturbances and modeling uncertainties to achieve the best transient performance. The proposed method is validated with simulation placed on a Furuta inverted pendulum, showing significant tracking performance improvement.

eess.SY↗
arXiv2023

On Robust Observer Design for System Motion on SE(3) Using Onboard Visual Sensors

Onboard visual sensing has been widely used in the unmanned ground vehicle (UGV) and/or unmanned aerial vehicle (UAV), which can be modeled as dynamic systems on SE(3). The onboard sensing outputs of the dynamic system can usually be applied to derive the relative position between the feature marks and the system, but bearing with explicit geometrical constraint. Such a visual geometrical constraint makes the design of the visual observer on SE(3) very challenging, as it will cause a time-varying or switching visible set due to the varying number of feature marks in this set along different trajectories. Moreover, the possibility of having mis-identified feature marks and modeling uncertainties might result in a divergent estimation error. This paper proposes a new robust observer design method that can accommodate these uncertainties from onboard visual sensing. The key design idea for this observer is to estimate the visible set and identify the mis-identified features from the measurements. Based on the identified uncertainties, a switching strategy is proposed to ensure bounded estimation error for any given trajectory over a fixed time interval. Simulation results are provided to demonstrate the effectiveness of the proposed robust observer.

eess.SY↗
arXiv2023

Learning A Sparse Transformer Network for Effective Image Deraining

Transformers-based methods have achieved significant performance in image deraining as they can model the non-local information which is vital for high-quality image reconstruction. In this paper, we find that most existing Transformers usually use all similarities of the tokens from the query-key pairs for the feature aggregation. However, if the tokens from the query are different from those of the key, the self-attention values estimated from these tokens also involve in feature aggregation, which accordingly interferes with the clear image restoration. To overcome this problem, we propose an effective DeRaining network, Sparse Transformer (DRSformer) that can adaptively keep the most useful self-attention values for feature aggregation so that the aggregated features better facilitate high-quality image reconstruction. Specifically, we develop a learnable top-k selection operator to adaptively retain the most crucial attention scores from the keys for each query for better feature aggregation. Simultaneously, as the naive feed-forward network in Transformers does not model the multi-scale information that is important for latent clear image restoration, we develop an effective mixed-scale feed-forward network to generate better features for image deraining. To learn an enriched set of hybrid features, which combines local context from CNN operators, we equip our model with mixture of experts feature compensator to present a cooperation refinement deraining scheme. Extensive experimental results on the commonly used benchmarks demonstrate that the proposed method achieves favorable performance against state-of-the-art approaches. The source code and trained models are available at https://github.com/cschenxiang/DRSformer.

cs.CV↗
arXiv2023

Multimodal Analogical Reasoning over Knowledge Graphs

Analogical reasoning is fundamental to human cognition and holds an important place in various fields. However, previous studies mainly focus on single-modal analogical reasoning and ignore taking advantage of structure knowledge. Notably, the research in cognitive psychology has demonstrated that information from multimodal sources always brings more powerful cognitive transfer than single modality sources. To this end, we introduce the new task of multimodal analogical reasoning over knowledge graphs, which requires multimodal reasoning ability with the help of background knowledge. Specifically, we construct a Multimodal Analogical Reasoning dataSet (MARS) and a multimodal knowledge graph MarKG. We evaluate with multimodal knowledge graph embedding and pre-trained Transformer baselines, illustrating the potential challenges of the proposed task. We further propose a novel model-agnostic Multimodal analogical reasoning framework with Transformer (MarT) motivated by the structure mapping theory, which can obtain better performance. Code and datasets are available in https://github.com/zjunlp/MKG_Analogy.

cs.CL↗
arXiv2023

Self-Supervised Interest Transfer Network via Prototypical Contrastive Learning for Recommendation

Cross-domain recommendation has attracted increasing attention from industry and academia recently. However, most existing methods do not exploit the interest invariance between domains, which would yield sub-optimal solutions. In this paper, we propose a cross-domain recommendation method: Self-supervised Interest Transfer Network (SITN), which can effectively transfer invariant knowledge between domains via prototypical contrastive learning. Specifically, we perform two levels of cross-domain contrastive learning: 1) instance-to-instance contrastive learning, 2) instance-to-cluster contrastive learning. Not only that, we also take into account users' multi-granularity and multi-view interests. With this paradigm, SITN can explicitly learn the invariant knowledge of interest clusters between domains and accurately capture users' intents and preferences. We conducted extensive experiments on a public dataset and a large-scale industrial dataset collected from one of the world's leading e-commerce corporations. The experimental results indicate that SITN achieves significant improvements over state-of-the-art recommendation methods. Additionally, SITN has been deployed on a micro-video recommendation platform, and the online A/B testing results further demonstrate its practical value. Supplement is available at: https://github.com/fanqieCoffee/SITN-Supplement.

cs.IR↗
arXiv2023

Binary Embedding-based Retrieval at Tencent

Large-scale embedding-based retrieval (EBR) is the cornerstone of search-related industrial applications. Given a user query, the system of EBR aims to identify relevant information from a large corpus of documents that may be tens or hundreds of billions in size. The storage and computation turn out to be expensive and inefficient with massive documents and high concurrent queries, making it difficult to further scale up. To tackle the challenge, we propose a binary embedding-based retrieval (BEBR) engine equipped with a recurrent binarization algorithm that enables customized bits per dimension. Specifically, we compress the full-precision query and document embeddings, formulated as float vectors in general, into a composition of multiple binary vectors using a lightweight transformation model with residual multilayer perception (MLP) blocks. We can therefore tailor the number of bits for different applications to trade off accuracy loss and cost savings. Importantly, we enable task-agnostic efficient training of the binarization model using a new embedding-to-embedding strategy. We also exploit the compatible training of binary embeddings so that the BEBR engine can support indexing among multiple embedding versions within a unified system. To further realize efficient search, we propose Symmetric Distance Calculation (SDC) to achieve lower response time than Hamming codes. We successfully employed the introduced BEBR to Tencent products, including Sogou, Tencent Video, QQ World, etc. The binarization algorithm can be seamlessly generalized to various tasks with multiple modalities. Extensive experiments on offline benchmarks and online A/B tests demonstrate the efficiency and effectiveness of our method, significantly saving 30%~50% index costs with almost no loss of accuracy at the system level.

cs.IR↗
arXiv2023

Adversarial Attacks against Windows PE Malware Detection: A Survey of the State-of-the-Art

Malware has been one of the most damaging threats to computers that span across multiple operating systems and various file formats. To defend against ever-increasing and ever-evolving malware, tremendous efforts have been made to propose a variety of malware detection that attempt to effectively and efficiently detect malware so as to mitigate possible damages as early as possible. Recent studies have shown that, on the one hand, existing ML and DL techniques enable superior solutions in detecting newly emerging and previously unseen malware. However, on the other hand, ML and DL models are inherently vulnerable to adversarial attacks in the form of adversarial examples. In this paper, we focus on malware with the file format of portable executable (PE) in the family of Windows operating systems, namely Windows PE malware, as a representative case to study the adversarial attack methods in such adversarial settings. To be specific, we start by first outlining the general learning framework of Windows PE malware detection based on ML/DL and subsequently highlighting three unique challenges of performing adversarial attacks in the context of Windows PE malware. Then, we conduct a comprehensive and systematic review to categorize the state-of-the-art adversarial attacks against PE malware detection, as well as corresponding defenses to increase the robustness of Windows PE malware detection. Finally, we conclude the paper by first presenting other related attacks against Windows PE malware detection beyond the adversarial attacks and then shedding light on future research directions and opportunities. In addition, a curated resource list of adversarial attacks and defenses for Windows PE malware detection is also available at https://github.com/ryderling/adversarial-attacks-and-defenses-for-windows-pe-malware-detection.

cs.CR↗
arXiv2023

Exotic single-photon and enhanced deep-level emissions in hBN strain superlattice

The peculiar defect-related photon emission processes in 2D hexagonal boron nitride (hBN) have become a topic of intense research due to their potential applications in quantum information and sensing technologies. Recent efforts have focused on activating and modulating the defect energy levels in hBN by methods that can be integrated on a chip, and understanding the underlying physical mechanism. Here, we report on exotic single photon and enhanced deep-level emissions in 2D hBN strain superlattice, which is fabricated by transferring multilayer hBN onto hexagonal close-packed silica spheres on silica substrate. We realize effective activation of the single photon emissions (SPEs) in the multilayer hBN at the positions that are in contact with the apex of the SiO2 spheres. At these points, the local tensile strain induced blue-shift of the SPE is found to be up to 12 nm. Furthermore, high spatial resolution cathodoluminescence measurments show remarkable strain-enhanced deep-level (DL) emissions in the multilayer hBN with the emission intensity distribution following the periodic hexagonal pattern of the strain superlattice. The maximum DL emission enhancement is up to 350% with a energy redshift of 6 nm. Our results provide a simple on-chip compatible method for activating and tuning the defect-related photon emissions in multilayer hBN, demonstrating the potential of hBN strain superlattice as a building block for future on-chip quantum nanophotonic devices.

cond-mat.mtrl-sci↗
arXiv2023

Perfect teleportation with a partially entangled quantum channel

Quantum teleportation provides a way to transfer unknown quantum states from one system to another via an entangled state as a quantum channel without physical transmission of the object itself. The entangled channel, measurement performed by the sender (Alice), and classical information sent to the receiver (Bob) are three key ingredients in the procedure, which need to cooperate with each other. To study the relationship among the three parts, we propose a scheme for perfect teleportation of a qubit through a high-dimensional quantum channel in a pure state with two equal largest Schmidt coefficients. The scheme requires less entanglement of Alice's measurement but more classical bits than the original scheme via a Bell state. The two quantities increase with the entanglement of the quantum channel when its dimension is fixed and thereby can be regard as Alice's necessary capabilities to use the quantum channel. And the two capabilities appear complementary to each other.

quant-ph↗
arXiv2022

How We Express Ourselves Freely: Censorship, Self-censorship, and Anti-censorship on a Chinese Social Media

Censorship, anti-censorship, and self-censorship in an authoritarian regime have been extensively studies, yet the relationship between these intertwined factors is not well understood. In this paper, we report results of a large-scale survey study (N = 526) with Sina Weibo users toward bridging this research gap. Through descriptive statistics, correlation analysis, and regression analysis, we uncover how users are being censored, how and why they conduct self-censorship on different topics and in different scenarios (i.e., post, repost, and comment), and their various anti-censorship strategies. We further identify the metrics of censorship and self-censorship, find the influence factors, and construct a mediation model to measure their relationship. Based on these findings, we discuss implications for democratic social media design and future censorship research.

cs.CY↗
arXiv2022

Joint segmentation and discontinuity-preserving deformable registration: Application to cardiac cine-MR images

Medical image registration is a challenging task involving the estimation of spatial transformations to establish anatomical correspondence between pairs or groups of images. Recently, deep learning-based image registration methods have been widely explored, and demonstrated to enable fast and accurate image registration in a variety of applications. However, most deep learning-based registration methods assume that the deformation fields are smooth and continuous everywhere in the image domain, which is not always true, especially when registering images whose fields of view contain discontinuities at tissue/organ boundaries. In such scenarios, enforcing smooth, globally continuous deformation fields leads to incorrect/implausible registration results. We propose a novel discontinuity-preserving image registration method to tackle this challenge, which ensures globally discontinuous and locally smooth deformation fields, leading to more accurate and realistic registration results. The proposed method leverages the complementary nature of image segmentation and registration and enables joint segmentation and pair-wise registration of images. A co-attention block is proposed in the segmentation component of the network to learn the structural correlations in the input images, while a discontinuity-preserving registration strategy is employed in the registration component of the network to ensure plausibility in the estimated deformation fields at tissue/organ interfaces. We evaluate our method on the task of intra-subject spatio-temporal image registration using large-scale cinematic cardiac magnetic resonance image sequences, and demonstrate that our method achieves significant improvements over the state-of-the-art for medical image registration, and produces high-quality segmentation masks for the regions of interest.

cs.CV↗
arXiv2022

On Analyzing the Role of Image for Visual-enhanced Relation Extraction

Multimodal relation extraction is an essential task for knowledge graph construction. In this paper, we take an in-depth empirical analysis that indicates the inaccurate information in the visual scene graph leads to poor modal alignment weights, further degrading performance. Moreover, the visual shuffle experiments illustrate that the current approaches may not take full advantage of visual information. Based on the above observation, we further propose a strong baseline with an implicit fine-grained multimodal alignment based on Transformer for multimodal relation extraction. Experimental results demonstrate the better performance of our method. Codes are available at https://github.com/zjunlp/DeepKE/tree/main/example/re/multimodal.

cs.CL↗