arXiv subjects

Wei Xia

Publications and source records attributed to Wei Xia.

At least 145 records · Page 8Linked to original sources

arXiv2022

Effective and Efficient Graph Learning for Multi-view Clustering

Despite the impressive clustering performance and efficiency in characterizing both the relationship between data and cluster structure, existing graph-based multi-view clustering methods still have the following drawbacks. They suffer from the expensive time burden due to both the construction of graphs and eigen-decomposition of Laplacian matrix, and fail to explore the cluster structure of large-scale data. Moreover, they require a post-processing to get the final clustering, resulting in suboptimal performance. Furthermore, rank of the learned view-consensus graph cannot approximate the target rank. In this paper, drawing the inspiration from the bipartite graph, we propose an effective and efficient graph learning model for multi-view clustering. Specifically, our method exploits the view-similar between graphs of different views by the minimization of tensor Schatten p-norm, which well characterizes both the spatial structure and complementary information embedded in graphs of different views. We learn view-consensus graph with adaptively weighted strategy and connectivity constraint such that the connected components indicates clusters directly. Our proposed algorithm is time-economical and obtains the stable results and scales well with the data size. Extensive experimental results indicate that our method is superior to state-of-the-art methods.

cs.LG↗

arXiv2021

Long Short-Term Transformer for Online Action Detection

We present Long Short-term TRansformer (LSTR), a temporal modeling algorithm for online action detection, which employs a long- and short-term memory mechanism to model prolonged sequence data. It consists of an LSTR encoder that dynamically leverages coarse-scale historical information from an extended temporal window (e.g., 2048 frames spanning of up to 8 minutes), together with an LSTR decoder that focuses on a short time window (e.g., 32 frames spanning 8 seconds) to model the fine-scale characteristics of the data. Compared to prior work, LSTR provides an effective and efficient method to model long videos with fewer heuristics, which is validated by extensive empirical analysis. LSTR achieves state-of-the-art performance on three standard online action detection benchmarks, THUMOS'14, TVSeries, and HACS Segment. Code has been made available at: https://xumingze0308.github.io/projects/lstr

cs.CV↗

arXiv2021

Deformations of Dolbeault cohomology classes

In this paper, we establish a deformation theory for Dolbeault cohomology classes valued in holomorphic tensor bundles. We prove the extension equation which will play the role of Maurer-Cartan equation. Following the classical theory of Kodaira-Spencer-Kuranishi, we construct a canonical complete family of deformations by using the power series method. We also prove a simple relation between the existence of deformations and the varying of the dimensions of Dolbeault cohomology. The deformations of $(p,q)$-forms is shown to be unobstructed under some mild conditions. By analyzing Nakamura's example of complex parallelizable manifolds, we will see that the deformation theory developed in this work provides precise explanations to the jumping phenomenon of Dolbeault cohomology.

math.DG↗

arXiv2021

Tunable intrinsic ferromagnetic topological phases in bulk van der Waals crystal MnSb6Te10

Intrinsic ferromagnetism is a crucial ingredient to realize quantum anomalous Hall effect in quasi two dimensional materials, thus the search of intrinsic ferromagnetic topological materials is one of the most concerned issues in the field of topological phases of matter. In this work, combining magnetotransport measurements, first principles calculations, and angle-resolved photoemission spectroscopy studies, we find that in MnSb6Te10, the n = 2 member of the MnSb2Te4/(Sb2Te3)n family, the strong magnetic competition realizes a fragile ferromagnetic ground state, which whereas easily enters into ferrimagnetic and the Z_2 antiferromagnetic topological insulator phase with warming to higher temperature. Interestingly, the system stays in an inversion-symmetry-protected axion insulator phase in the ferromagnetic ground state as well as in the external magnetic field driven spin-polarized FM phase and can be converted into a Weyl semimetal with multiple Weyl nodes in the valence bands with hole doping, which are manifested by the measured notable intrinsic anomalous Hall effect. Our work thus provides an intrinsic magnetic topological material which is highly tunable into versatile topological phases by temperature, magnetic field, as well as carrier doping.

cond-mat.mtrl-sci↗

arXiv2021

An Automatic Detection Method Of Cerebral Aneurysms In Time-Of-Flight Magnetic Resonance Angiography Images Based On Attention 3D U-Net

Background:Subarachnoid hemorrhage caused by ruptured cerebral aneurysm often leads to fatal consequences.However,if the aneurysm can be found and treated during asymptomatic periods,the probability of rupture can be greatly reduced.At present,time-of-flight magnetic resonance angiography is one of the most commonly used non-invasive screening techniques for cerebral aneurysm,and the application of deep learning technology in aneurysm detection can effectively improve the screening effect of aneurysm.Existing studies have found that three-dimensional features play an important role in aneurysm detection,but they require a large amount of training data and have problems such as a high false positive rate. Methods:This paper proposed a novel method for aneurysm detection.First,a fully automatic cerebral artery segmentation algorithm without training data was used to extract the volume of interest,and then the 3D U-Net was improved by the 3D SENet module to establish an aneurysm detection model.Eventually a set of fully automated,end-to-end aneurysm detection methods have been formed. Results:A total of 231 magnetic resonance angiography image data were used in this study,among which 132 were training sets,34 were internal test sets and 65 were external test sets.The presented method obtained 97.89% sensitivity in the five-fold cross-validation and obtained 91.0% sensitivity with 2.48 false positives/case in the detection of the external test sets. Conclusions:Compared with the results of our previous studies and other studies,the method in this paper achieves a very competitive sensitivity with less training data and maintains a low false positive rate.As the only method currently using 3D U-Net for aneurysm detection,it proves the feasibility and superior performance of this network in aneurysm detection,and also explores the potential of the channel attention mechanism in this task.

eess.IV↗

arXiv2021

Self-supervised Contrastive Attributed Graph Clustering

Attributed graph clustering, which learns node representation from node attribute and topological graph for clustering, is a fundamental but challenging task for graph analysis. Recently, methods based on graph contrastive learning (GCL) have obtained impressive clustering performance on this task. Yet, we observe that existing GCL-based methods 1) fail to benefit from imprecise clustering labels; 2) require a post-processing operation to get clustering labels; 3) cannot solve out-of-sample (OOS) problem. To address these issues, we propose a novel attributed graph clustering network, namely Self-supervised Contrastive Attributed Graph Clustering (SCAGC). In SCAGC, by leveraging inaccurate clustering labels, a self-supervised contrastive loss, which aims to maximize the similarities of intra-cluster nodes while minimizing the similarities of inter-cluster nodes, are designed for node representation learning. Meanwhile, a clustering module is built to directly output clustering labels by contrasting the representation of different clusters. Thus, for the OOS nodes, SCAGC can directly calculate their clustering labels. Extensive experimental results on four benchmark datasets have shown that SCAGC consistently outperforms 11 competitive clustering methods.

cs.LG↗

arXiv2021

Magneto-Seebeck effect and ambipolar Nernst effect in CsV$_3$Sb$_5$ superconductor

We present a study of Seebeck and Nernst effect in combination with magnetoresistance and Hall measurements of the Kagome superconductor CsV$_3$Sb$_5$. Sizable magneto-Seebeck signal appears once the charge density wave (CDW) order sets in below $T_{CDW}$=94 K. The Nernst signal peaks at a lower temperature $T^*$=35 K, crossing which the Hall coefficient switches sign, which we attribute to the ambipolar transport of compensated bands due to the multi-band nature of CsV$_3$Sb$_5$. Sublinear Nernst signal as a function of magnetic field, together with large anomalous Nernst effect (ANE) also emerge inside the CDW phase, despite the absence of long-range magnetic order. These findings suggest that, the transport properties are dominated by small pockets with multi-band profile, and that the unusual band topology also plays an import role in the CDW state of CsV$_3$Sb$_5$.

cond-mat.supr-con↗

arXiv2021

Scenario Aware Speech Recognition: Advancements for Apollo Fearless Steps & CHiME-4 Corpora

In this study, we propose to investigate triplet loss for the purpose of an alternative feature representation for ASR. We consider a general non-semantic speech representation, which is trained with a self-supervised criteria based on triplet loss called TRILL, for acoustic modeling to represent the acoustic characteristics of each audio. This strategy is then applied to the CHiME-4 corpus and CRSS-UTDallas Fearless Steps Corpus, with emphasis on the 100-hour challenge corpus which consists of 5 selected NASA Apollo-11 channels. An analysis of the extracted embeddings provides the foundation needed to characterize training utterances into distinct groups based on acoustic distinguishing properties. Moreover, we also demonstrate that triplet-loss based embedding performs better than i-Vector in acoustic modeling, confirming that the triplet loss is more effective than a speaker feature. With additional techniques such as pronunciation and silence probability modeling, plus multi-style training, we achieve a +5.42% and +3.18% relative WER improvement for the development and evaluation sets of the Fearless Steps Corpus. To explore generalization, we further test the same technique on the 1 channel track of CHiME-4 and observe a +11.90% relative WER improvement for real test data.

cs.SD↗

arXiv2021

Harnessing Unrecognizable Faces for Improving Face Recognition

The common implementation of face recognition systems as a cascade of a detection stage and a recognition or verification stage can cause problems beyond failures of the detector. When the detector succeeds, it can detect faces that cannot be recognized, no matter how capable the recognition system. Recognizability, a latent variable, should therefore be factored into the design and implementation of face recognition systems. We propose a measure of recognizability of a face image that leverages a key empirical observation: an embedding of face images, implemented by a deep neural network trained using mostly recognizable identities, induces a partition of the hypersphere whereby unrecognizable identities cluster together. This occurs regardless of the phenomenon that causes a face to be unrecognizable, it be optical or motion blur, partial occlusion, spatial quantization, poor illumination. Therefore, we use the distance from such an "unrecognizable identity" as a measure of recognizability, and incorporate it in the design of the over-all system. We show that accounting for recognizability reduces error rate of single-image face recognition by 58% at FAR=1e-5 on the IJB-C Covariate Verification benchmark, and reduces verification error rate by 24% at FAR=1e-5 in set-based recognition on the IJB-C benchmark.

cs.CV↗

arXiv2021

Deformations of Dolbeault cohomology classes for Lie algebra with complex structures

In this paper, we study deformations of complex structures on Lie algebras and its associated deformations of Dolbeault cohomology classes. A complete deformation of complex structures is constructed in a way similar to the Kuranishi family. The extension isomorphism is shown to be valid in this case. As an application, we prove that given a family of left invariant deformations $\{M_t\}_{t\in B}$ of a compact complex manifold $M=(Γ\setminus G, J)$ where $G$ is a Lie group, $Γ$ a sublattice and $J$ a left invariant complex structure, the set of all $t\in B$ such that the Dolbeault cohomology on $M_t$ may be computed by left invariant tensor fields is an analytic open subset of $B$.

math.DG↗

arXiv2021

Learning Self-Consistency for Deepfake Detection

We propose a new method to detect deepfake images using the cue of the source feature inconsistency within the forged images. It is based on the hypothesis that images' distinct source features can be preserved and extracted after going through state-of-the-art deepfake generation processes. We introduce a novel representation learning approach, called pair-wise self-consistency learning (PCL), for training ConvNets to extract these source features and detect deepfake images. It is accompanied by a new image synthesis approach, called inconsistency image generator (I2G), to provide richly annotated training data for PCL. Experimental results on seven popular datasets show that our models improve averaged AUC over the state of the art from 96.45% to 98.05% in the in-dataset evaluation and from 86.03% to 92.18% in the cross-dataset evaluation.

cs.CV↗

arXiv2021

Learning Hierarchical Graph Neural Networks for Image Clustering

We propose a hierarchical graph neural network (GNN) model that learns how to cluster a set of images into an unknown number of identities using a training set of images annotated with labels belonging to a disjoint set of identities. Our hierarchical GNN uses a novel approach to merge connected components predicted at each level of the hierarchy to form a new graph at the next level. Unlike fully unsupervised hierarchical clustering, the choice of grouping and complexity criteria stems naturally from supervision in the training set. The resulting method, Hi-LANDER, achieves an average of 54% improvement in F-score and 8% increase in Normalized Mutual Information (NMI) relative to current GNN-based clustering algorithms. Additionally, state-of-the-art GNN-based methods rely on separate models to predict linkage probabilities and node densities as intermediate steps of the clustering process. In contrast, our unified framework achieves a seven-fold decrease in computational cost. We release our training and inference code at https://github.com/dmlc/dgl/tree/master/examples/pytorch/hilander.

cs.CV↗

arXiv2021

Optical manipulation of electronic dimensionality in a quantum material

Exotic phenomenon can be achieved in quantum materials by confining electronic states into two dimensions. For example, relativistic fermions are realised in a single layer of carbon atoms, the quantized Hall effect can result from two-dimensional (2D) systems, and the superconducting transition temperature can be enhanced significantly in a one-atomic-layer material. Ordinarily, 2D electronic system can be obtained by exfoliating the layered materials, growing monolayer materials on substrates, or establishing interfaces between different materials. Herein, we use femtosecond infrared laser pulses to invert the periodic lattice distortion sectionally in a three-dimensional (3D) charge density wave material, creating macroscopic domain walls of transient 2D ordered electronic states with exotic properties. The corresponding ultrafast electronic and lattice dynamics are captured by time- and angle-resolved photoemission spectroscopy and MeV ultrafast electron diffraction. Surprisingly, a novel energy gap state, which might be a signature of light-induced superconductivity, is identified in the photoinduced 2D domain wall near the surface. Such optical modulation of atomic motion is a new path to realise 2D electronic states and will be a new platform for creating novel phases in quantum materials.

cond-mat.str-el↗

arXiv2021

Semi-TCL: Semi-Supervised Track Contrastive Representation Learning

Online tracking of multiple objects in videos requires strong capacity of modeling and matching object appearances. Previous methods for learning appearance embedding mostly rely on instance-level matching without considering the temporal continuity provided by videos. We design a new instance-to-track matching objective to learn appearance embedding that compares a candidate detection to the embedding of the tracks persisted in the tracker. It enables us to learn not only from videos labeled with complete tracks, but also unlabeled or partially labeled videos. We implement this learning objective in a unified form following the spirit of constrastive loss. Experiments on multiple object tracking datasets demonstrate that our method can effectively learning discriminative appearance embeddings in a semi-supervised fashion and outperform state of the art methods on representative benchmarks.

cs.CV↗

arXiv2021

On the deformed Bott-Chern cohomology

Given a compact complex manifold $X$ and a integrable Beltrami differential $ϕ\in A^{0,1}(X, T_{X}^{1,0})$, we introduce a double complex structure on $A^{\bullet,\bullet}(X)$ naturally determined by $ϕ$ and study its Bott-Chern cohomology. In particular, we establish a deformation theory for Bott-Chern cohomology and use it to compute the deformed Bott-Chern cohomology for the Iwasawa manifold and the holomorphically parallelizable Nakamura manifold. The $\partial\bar{\partial}_ϕ$-lemma is studied and we show a compact complex manifold satisfying $\partial\bar{\partial}_ϕ$-lemma is formal.

math.DG↗

arXiv2021

Positive-Congruent Training: Towards Regression-Free Model Updates

Reducing inconsistencies in the behavior of different versions of an AI system can be as important in practice as reducing its overall error. In image classification, sample-wise inconsistencies appear as "negative flips": A new model incorrectly predicts the output for a test sample that was correctly classified by the old (reference) model. Positive-congruent (PC) training aims at reducing error rate while at the same time reducing negative flips, thus maximizing congruency with the reference model only on positive predictions, unlike model distillation. We propose a simple approach for PC training, Focal Distillation, which enforces congruence with the reference model by giving more weights to samples that were correctly classified. We also found that, if the reference model itself can be chosen as an ensemble of multiple deep neural networks, negative flips can be further reduced without affecting the new model's accuracy.

cs.CV↗

arXiv2021

Compatibility-aware Heterogeneous Visual Search

We tackle the problem of visual search under resource constraints. Existing systems use the same embedding model to compute representations (embeddings) for the query and gallery images. Such systems inherently face a hard accuracy-efficiency trade-off: the embedding model needs to be large enough to ensure high accuracy, yet small enough to enable query-embedding computation on resource-constrained platforms. This trade-off could be mitigated if gallery embeddings are generated from a large model and query embeddings are extracted using a compact model. The key to building such a system is to ensure representation compatibility between the query and gallery models. In this paper, we address two forms of compatibility: One enforced by modifying the parameters of each model that computes the embeddings. The other by modifying the architectures that compute the embeddings, leading to compatibility-aware neural architecture search (CMP-NAS). We test CMP-NAS on challenging retrieval tasks for fashion images (DeepFashion2), and face images (IJB-C). Compared to ordinary (homogeneous) visual search using the largest embedding model (paragon), CMP-NAS achieves 80-fold and 23-fold cost reduction while maintaining accuracy within 0.3% and 1.6% of the paragon on DeepFashion2 and IJB-C respectively.

cs.CV↗

arXiv2021

Emergence of new van Hove singularities in the charge density wave state of a topological kagome metal RbV3Sb5

Quantum materials with layered kagome structures have drawn considerable attention due to their unique lattice geometry, which gives rise to flat bands co-existing with Dirac-like dispersions. The interplay between strong Coulomb correlations and nontrivial band topology in these systems results in various exotic phenomena. Recently, vanadium-based materials with layered kagome structures are discovered to be topological metals, which exhibit charge density wave (CDW) properties, significant anomalous Hall effect, and unusual superconductivity at low temperatures. Here we exploit high-resolution angle-resolved photoemission spectroscopy to investigate the electronic structure evolution induced by the CDW transition in a vanadium-based kagome material RbV3Sb5. A remarkable band renormalization in the CDW state is observed, which is consistent with first principles calculations based on an inverse star-of-David superstructure. The CDW phase transition gives rise to a partial energy gap opening at the Fermi level, a shift in the band dispersion, and most importantly, the emergence of new van Hove singularities associated with large density of states, which are absent in the normal phase and may be related to superconductivity observed at lower temperatures. Our work would shed light on the microscopic mechanisms for the formation of the CDW and superconducting states in these topological kagome metals.

cond-mat.supr-con↗