arXiv subjects

Ke Wang

Publications and source records attributed to Ke Wang.

At least 397 records · Page 22Linked to original sources

arXiv2023

Axion Insulator State in Hundred-Nanometer-Thick Magnetic Topological Insulator Sandwich Heterostructures

An axion insulator is a three-dimensional (3D) topological insulator (TI), in which the bulk maintains the time-reversal symmetry or inversion symmetry but the surface states are gapped by surface magnetization. The axion insulator state has been observed in molecular beam epitaxy (MBE)-grown magnetically doped TI sandwiches and exfoliated intrinsic magnetic TI MnBi2Te4 flakes with an even number layer. All these samples have a thickness of ~10 nm, near the 2D-to-3D boundary. The coupling between the top and bottom surface states in thin samples may hinder the observation of quantized topological magnetoelectric response. Here, we employ MBE to synthesize magnetic TI sandwich heterostructures and find that the axion insulator state persists in a 3D sample with a thickness of ~106 nm. Our transport results show that the axion insulator state starts to emerge when the thickness of the middle undoped TI layer is greater than ~3 nm. The 3D hundred-nanometer-thick axion insulator provides a promising platform for the exploration of the topological magnetoelectric effect and other emergent magnetic topological states, such as the high-order TI phase.

cond-mat.mes-hall↗

arXiv2023

The ALMA Survey of Star Formation and Evolution in Massive Protoclusters with Blue Profiles (ASSEMBLE): Core Growth, Cluster Contraction, and Primordial Mass Segregation

The ALMA Survey of Star Formation and Evolution in Massive Protoclusters with Blue Profiles (ASSEMBLE) aims to investigate the process of mass assembly and its connection to high-mass star formation theories in protoclusters in a dynamic view. We observed 11 massive (Mclump>1000 Msun), luminous (Lbol>10,000 Lsun), and blue-profile (infall signature) clumps by ALMA with resolution of 2200-5500 au at 350 GHz (870 um) in continuum and line emission. 248 dense cores were identified, including 106 cores showing protostellar signatures and 142 prestellar core candidates. Compared to early-stage infrared dark clouds (IRDCs) by ASHES, the core mass and surface density within the ASSEMBLE clumps exhibited significant increment, suggesting concurrent core accretion during the evolution of the clumps. The maximum mass of prestellar cores was found to be 2 times larger than that in IRDCs, indicating evolved protoclusters have the potential to harbor massive prestellar cores. The mass relation between clumps and their most massive core (MMCs) is observed in ASSEMBLE but not in IRDCs, which is suggested to be regulated by multiscale mass accretion. The mass correlation between the core clusters and their MMCs has a steeper slope compared to that observed in stellar clusters, which can be due to fragmentation of the MMC and stellar multiplicity. We observe a decrease in core separation and an increase in central concentration as protoclusters evolve. We confirm primordial mass segregation in the ASSEMBLE protoclusters, possibly resulting from gravitational concentration and/or gas accretion.

astro-ph.GA↗

arXiv2023

Clump-scale Gas Infall in High-mass Star Formation: a Multi-transition View with JCMT HCN (4--3) Mapping

Gas infall motions play a crucial role in high-mass star formation and are characterized by observable signatures in the form of blue-shifted asymmetric spectral line profiles ("blue profiles"). However, the connection between blue profiles and infall motions is unclear due to complex gas motions at parsec scales. In this study, we present the results of an HCN (4-3) mapping survey conducted with the JCMT, towards 38 massive clumps exhibiting blue profiles in HCO+ (3-2). We extract 34 HCN cores from the 38 observed fields. The core-averaged spectra show various line profiles, indicating that blue-profile HCO+ (3-2) does not guarantee the same in HCN (4-3). Through non-LTE radiation transfer calculations, we attribute the low detection rate of high-$J$ blue profiles to a combination of insufficient HCN (4-3) opacity and intricate gas motion across different density layers. The comparison between the MALT90 and BGPS line surveys highlights the importance of appropriate tracers, high spectral resolution, and column density thresholds when searching for blue profiles. We select 11 reliable infall candidates and adopt the Hill5 model to fit the infall velocity of 0.2-1.9 km/s, corresponding to 5% to 74% of free-fall velocity. Assuming a spherically collapsing model, we estimate the median and mean mass infall rates to be 4.5E-3 and 7.6E-3 Msun/year, respectively. The consistency of the mass infall rates among different transitions suggests a steady accretion process from the clump gas envelope to the inner region.

astro-ph.GA↗

arXiv2023

Constraints on the Local Cosmic Void from the Pantheon Supernovae Data

In principle, the local cosmic void can be simply modeled by the spherically symmetric Lemaitre-Tolman-Bondi (LTB) metric. In practice, the real local cosmic void is probably not spherically symmetric. In this paper, to reconstruct a more realistic profile of the local cosmic void, we divide it into several segments. Each segment with certain solid angle is modeled by its own LTB metric. Meanwhile, we divide the 1048 type Ia supernovae (SNIa) of the Pantheon Survey into corresponding subsets according to their distribution in the galactic coordinate system. Obviously, each SNIa subset can only be used to reconstruct the profile of one segment. Finally, we can patch together an irregular profile for the local cosmic void with the whole Pantheon sample. Note that, the paucity of each data subset lead us to focus on the inner part of each void segment and assume that the half radii of the void segments are sufficient to constrain the whole segment. We find that, despite $2σ$ signals of anisotropy limited to the depth of the void segments, the constraints on every void segment are consistent with $Λ$CDM model at $95\%$ CL. Moreover, our constraints are too weak to challenge the cosmic homogeneity and isotropy.

astro-ph.CO↗

arXiv2023

Artificial Intelligence Techniques for Next-Generation Mega Satellite Networks

Space communications, particularly massive satellite networks, re-emerged as an appealing candidate for next generation networks due to major advances in space launching, electronics, processing power, and miniaturization. However, massive satellite networks rely on numerous underlying and intertwined processes that cannot be truly captured using conventionally used models, due to their dynamic and unique features such as orbital speed, inter-satellite links, short pass time, and satellite footprint, among others. Hence, new approaches are needed to enable the network to proactively adjust to the rapidly varying conditions associated within the link. Artificial intelligence (AI) provides a pathway to capture these processes, analyze their behavior, and model their effect on the network. This article introduces the application of AI techniques for integrated terrestrial satellite networks, particularly massive satellite network communications. It details the unique features of massive satellite networks, and the overarching challenges concomitant with their integration into the current communication infrastructure. Moreover, this article provides insights into state-of-the-art AI techniques across various layers of the communication link. This entails applying AI for forecasting the highly dynamic radio channel, spectrum sensing and classification, signal detection and demodulation, inter-satellite and satellite access network optimization, and network security. Moreover, future paradigms and the mapping of these mechanisms onto practical networks are outlined.

cs.NI↗

arXiv2023

Computing SHAP Efficiently Using Model Structure Information

SHAP (SHapley Additive exPlanations) has become a popular method to attribute the prediction of a machine learning model on an input to its features. One main challenge of SHAP is the computation time. An exact computation of Shapley values requires exponential time complexity. Therefore, many approximation methods are proposed in the literature. In this paper, we propose methods that can compute SHAP exactly in polynomial time or even faster for SHAP definitions that satisfy our additivity and dummy assumptions (eg, kernal SHAP and baseline SHAP). We develop different strategies for models with different levels of model structure information: known functional decomposition, known order of model (defined as highest order of interaction in the model), or unknown order. For the first case, we demonstrate an additive property and a way to compute SHAP from the lower-order functional components. For the second case, we derive formulas that can compute SHAP in polynomial time. Both methods yield exact SHAP results. Finally, if even the order of model is unknown, we propose an iterative way to approximate Shapley values. The three methods we propose are computationally efficient when the order of model is not high which is typically the case in practice. We compare with sampling approach proposed in Castor & Gomez (2008) using simulation studies to demonstrate the efficacy of our proposed methods.

stat.ML↗

arXiv2023

RestGPT: Connecting Large Language Models with Real-World RESTful APIs

Tool-augmented large language models (LLMs) have achieved remarkable progress in tackling a broad range of tasks. However, existing methods are mainly restricted to specifically designed tools and fail to fulfill complex instructions, having great limitations when confronted with real-world scenarios. In this paper, we explore a more realistic scenario by connecting LLMs with RESTful APIs, which adhere to the widely adopted REST software architectural style for web service development. To address the practical challenges of tackling complex instructions, we propose RestGPT, which exploits the power of LLMs and conducts a coarse-to-fine online planning mechanism to enhance the abilities of task decomposition and API selection. RestGPT also contains an API executor tailored for calling RESTful APIs, which can meticulously formulate parameters and parse API responses. To fully evaluate the performance of RestGPT, we propose RestBench, a high-quality benchmark which consists of two real-world scenarios and human-annotated instructions with gold solution paths. Experiments show that RestGPT is able to achieve impressive results in complex tasks and has strong robustness, which paves a new way towards AGI. RestGPT and RestBench is publicly available at https://restgpt.github.io/.

cs.CL↗

arXiv2023

OFVL-MS: Once for Visual Localization across Multiple Indoor Scenes

In this work, we seek to predict camera poses across scenes with a multi-task learning manner, where we view the localization of each scene as a new task. We propose OFVL-MS, a unified framework that dispenses with the traditional practice of training a model for each individual scene and relieves gradient conflict induced by optimizing multiple scenes collectively, enabling efficient storage yet precise visual localization for all scenes. Technically, in the forward pass of OFVL-MS, we design a layer-adaptive sharing policy with a learnable score for each layer to automatically determine whether the layer is shared or not. Such sharing policy empowers us to acquire task-shared parameters for a reduction of storage cost and task-specific parameters for learning scene-related features to alleviate gradient conflict. In the backward pass of OFVL-MS, we introduce a gradient normalization algorithm that homogenizes the gradient magnitude of the task-shared parameters so that all tasks converge at the same pace. Furthermore, a sparse penalty loss is applied on the learnable scores to facilitate parameter sharing for all tasks without performance degradation. We conduct comprehensive experiments on multiple benchmarks and our new released indoor dataset LIVL, showing that OFVL-MS families significantly outperform the state-of-the-arts with fewer parameters. We also verify that OFVL-MS can generalize to a new scene with much few parameters while gaining superior localization performance.

cs.CV↗

arXiv2023

SwinFace: A Multi-task Transformer for Face Recognition, Expression Recognition, Age Estimation and Attribute Estimation

In recent years, vision transformers have been introduced into face recognition and analysis and have achieved performance breakthroughs. However, most previous methods generally train a single model or an ensemble of models to perform the desired task, which ignores the synergy among different tasks and fails to achieve improved prediction accuracy, increased data efficiency, and reduced training time. This paper presents a multi-purpose algorithm for simultaneous face recognition, facial expression recognition, age estimation, and face attribute estimation (40 attributes including gender) based on a single Swin Transformer. Our design, the SwinFace, consists of a single shared backbone together with a subnet for each set of related tasks. To address the conflicts among multiple tasks and meet the different demands of tasks, a Multi-Level Channel Attention (MLCA) module is integrated into each task-specific analysis subnet, which can adaptively select the features from optimal levels and channels to perform the desired tasks. Extensive experiments show that the proposed model has a better understanding of the face and achieves excellent performance for all tasks. Especially, it achieves 90.97% accuracy on RAF-DB and 0.22 $ε$-error on CLAP2015, which are state-of-the-art results on facial expression recognition and age estimation respectively. The code and models will be made publicly available at https://github.com/lxq1000/SwinFace.

cs.CV↗

arXiv2023

Solving Challenging Math Word Problems Using GPT-4 Code Interpreter with Code-based Self-Verification

Recent progress in large language models (LLMs) like GPT-4 and PaLM-2 has brought significant advancements in addressing math reasoning problems. In particular, OpenAI's latest version of GPT-4, known as GPT-4 Code Interpreter, shows remarkable performance on challenging math datasets. In this paper, we explore the effect of code on enhancing LLMs' reasoning capability by introducing different constraints on the \textit{Code Usage Frequency} of GPT-4 Code Interpreter. We found that its success can be largely attributed to its powerful skills in generating and executing code, evaluating the output of code execution, and rectifying its solution when receiving unreasonable outputs. Based on this insight, we propose a novel and effective prompting method, explicit \uline{c}ode-based \uline{s}elf-\uline{v}erification~(CSV), to further boost the mathematical reasoning potential of GPT-4 Code Interpreter. This method employs a zero-shot prompt on GPT-4 Code Interpreter to encourage it to use code to self-verify its answers. In instances where the verification state registers as ``False'', the model shall automatically amend its solution, analogous to our approach of rectifying errors during a mathematics examination. Furthermore, we recognize that the states of the verification result indicate the confidence of a solution, which can improve the effectiveness of majority voting. With GPT-4 Code Interpreter and CSV, we achieve an impressive zero-shot accuracy on MATH dataset \textbf{(53.9\% $\to$ 84.3\%)}.

cs.CL↗

arXiv2023

Simulating Cosmological Evolution by Quantum Quench of an Atomic BEC

In cosmological evolution, it is the homogeneous scalar field (inflaton) that drives the universe to expand isotropically and to generate standard model particles. However, to simulate cosmology, atomic gas research has focused on the dynamics of Bose-Einstein condensates (BEC) with continuously applied forces. In this paper we argue a complementary approach needs also to be pursued; we, thus, consider the analogue BEC experiments in a non-driven, closed atomic system. We implement this using a BEC in an optical lattice which, after a quench, freely transitions from an unstable to a stable state. This dynamical evolution displays the counterpart "preheating", "reheating" and "thermalization" phases of cosmology. Importantly, our studies of these analogue processes yield tractable analytic models. Of great utility to the cold atom community, such understanding elucidates the dynamics of non-adiabatic condensate preparation.

cond-mat.quant-gas↗

arXiv2023

Poly-MOT: A Polyhedral Framework For 3D Multi-Object Tracking

3D Multi-object tracking (MOT) empowers mobile robots to accomplish well-informed motion planning and navigation tasks by providing motion trajectories of surrounding objects. However, existing 3D MOT methods typically employ a single similarity metric and physical model to perform data association and state estimation for all objects. With large-scale modern datasets and real scenes, there are a variety of object categories that commonly exhibit distinctive geometric properties and motion patterns. In this way, such distinctions would enable various object categories to behave differently under the same standard, resulting in erroneous matches between trajectories and detections, and jeopardizing the reliability of downstream tasks (navigation, etc.). Towards this end, we propose Poly-MOT, an efficient 3D MOT method based on the Tracking-By-Detection framework that enables the tracker to choose the most appropriate tracking criteria for each object category. Specifically, Poly-MOT leverages different motion models for various object categories to characterize distinct types of motion accurately. We also introduce the constraint of the rigid structure of objects into a specific motion model to accurately describe the highly nonlinear motion of the object. Additionally, we introduce a two-stage data association strategy to ensure that objects can find the optimal similarity metric from three custom metrics for their categories and reduce missing matches. On the NuScenes dataset, our proposed method achieves state-of-the-art performance with 75.4\% AMOTA. The code is available at https://github.com/lixiaoyu2000/Poly-MOT

cs.RO↗

arXiv2023

Quantum $z=2$ Lifshitz criticality in one-dimensional interacting fermions

We consider Lifshitz criticality (LC) with the dynamical critical exponent $z=2$ in one-dimensional interacting fermions with a filled Dirac Sea. We report that interactions have crucial effects on Lifshitz criticality. Single particle excitations are destabilized by interaction and decay into the particle-hole continuum, which is reflected in the logarithmic divergence in the imaginary part of one-loop self-energy. We show that the system is sensitive to the sign of interaction. Random-phase approximation (RPA) shows that the collective particle-hole excitations emerge only when the interaction is repulsive. The dispersion of collective modes is gapless and linear. If the interaction is attractive, the one-loop renormalization group (RG) shows that there may exist a stable RG fixed point described by two coupling constants. We also show that the on-site interaction (without any other perturbations at the UV scale) would always turn on the relevant velocity perturbation to the quadratic Lagrangian in the RG flow, driving the system flow to the conformal-invariant criticality. In the numerical simulations of the lattice model at the half-filling, we find that, for either on-site positive or negative interactions, the dynamical critical exponent becomes $z=1$ in the infrared (IR) limit and the entanglement entropy is a logarithmic function of the system size $L$. The work paves the way to study one-dimensional interacting LCs.

cond-mat.str-el↗

arXiv2023

Existence and regularity of pullback attractors for nonclassical non-autonomous diffusion equations with delay

In this paper, we consider the asymptotic behavior of weak solutions for non-autonomous diffusion equations with delay in time-dependent spaces when the nonlinear function $f$ is critical growth, the delay term $g(t, u_t)$ contains some hereditary characteristics and the external force $h \in L_{l o c}^{2}\left(\mathbb{R} ; L^{2}(Ω)\right)$. Firstly, we prove the well-posedness of solutions by using the Faedo-Galerkin approximation method. Then after a series of elaborate energy estimates and calculations, we establish the existence and regularity of pullback attractors in time-dependent spaces $C_{\mathcal{H}_{t}(Ω)}$ and $C_{\mathcal{H}^{1}_{t}(Ω)}$ respectively.

math.AP↗

arXiv2023

Photochemical origin of SiC$_2$ in the circumstellar envelope of carbon-rich AGB stars revealed by ALMA

Whether SiC$_2$ is a parent species, that is formed in the photosphere or as a by-product of high-temperature dust formation, or a daughter species, formed in a chemistry driven by the photodestruction of parent species in the outer envelope, has been debated for a long time. Here, we analyze the ALMA observations of four SiC$_2$ transitions in the CSEs of three C-rich AGB stars (AI Vol, II Lup, and RAFGL 4211), and found that SiC$_2$ exhibits an annular, shell-like distribution in these targets, suggesting that SiC$_2$ can be a daughter species in the CSEs of carbon-rich AGB stars. The results can provide important references for future chemical models.

astro-ph.SR↗

arXiv2023

Learning Gaussian Graphical Models with Latent Confounders

Gaussian Graphical models (GGM) are widely used to estimate the network structures in many applications ranging from biology to finance. In practice, data is often corrupted by latent confounders which biases inference of the underlying true graphical structure. In this paper, we compare and contrast two strategies for inference in graphical models with latent confounders: Gaussian graphical models with latent variables (LVGGM) and PCA-based removal of confounding (PCA+GGM). While these two approaches have similar goals, they are motivated by different assumptions about confounding. In this paper, we explore the connection between these two approaches and propose a new method, which combines the strengths of these two approaches. We prove the consistency and convergence rate for the PCA-based method and use these results to provide guidance about when to use each method. We demonstrate the effectiveness of our methodology using both simulations and in two real-world applications.

stat.ME↗

arXiv2023

Benign Overfitting in Multiclass Classification: All Roads Lead to Interpolation

The literature on "benign overfitting" in overparameterized models has been mostly restricted to regression or binary classification; however, modern machine learning operates in the multiclass setting. Motivated by this discrepancy, we study benign overfitting in multiclass linear classification. Specifically, we consider the following training algorithms on separable data: (i) empirical risk minimization (ERM) with cross-entropy loss, which converges to the multiclass support vector machine (SVM) solution; (ii) ERM with least-squares loss, which converges to the min-norm interpolating (MNI) solution; and, (iii) the one-vs-all SVM classifier. First, we provide a simple sufficient deterministic condition under which all three algorithms lead to classifiers that interpolate the training data and have equal accuracy. When the data is generated from Gaussian mixtures or a multinomial logistic model, this condition holds under high enough effective overparameterization. We also show that this sufficient condition is satisfied under "neural collapse", a phenomenon that is observed in training deep neural networks. Second, we derive novel bounds on the accuracy of the MNI classifier, thereby showing that all three training algorithms lead to benign overfitting under sufficient overparameterization. Ultimately, our analysis shows that good generalization is possible for SVM solutions beyond the realm in which typical margin-based bounds apply.

stat.ML↗

arXiv2023

ECL: Class-Enhancement Contrastive Learning for Long-tailed Skin Lesion Classification

Skin image datasets often suffer from imbalanced data distribution, exacerbating the difficulty of computer-aided skin disease diagnosis. Some recent works exploit supervised contrastive learning (SCL) for this long-tailed challenge. Despite achieving significant performance, these SCL-based methods focus more on head classes, yet ignoring the utilization of information in tail classes. In this paper, we propose class-Enhancement Contrastive Learning (ECL), which enriches the information of minority classes and treats different classes equally. For information enhancement, we design a hybrid-proxy model to generate class-dependent proxies and propose a cycle update strategy for parameters optimization. A balanced-hybrid-proxy loss is designed to exploit relations between samples and proxies with different classes treated equally. Taking both "imbalanced data" and "imbalanced diagnosis difficulty" into account, we further present a balanced-weighted cross-entropy loss following curriculum learning schedule. Experimental results on the classification of imbalanced skin lesion data have demonstrated the superiority and effectiveness of our method.

cs.CV↗