arXiv subjects

Xiao Wang

Publications and source records attributed to Xiao Wang.

At least 595 records · Page 33Linked to original sources

arXiv2021

Aligned Stacking of Nanopatterned 2D materials -- Towards 3D printing at atomic resolution

Two-dimensional materials can be combined by placing individual layers on top of each other, so that they are bound only by their van der Waals interaction. The sequence of layers can be chosen arbitrarily, enabling an essentially atomic-level control of the material and thereby a wide choice of properties along one dimension. However, simultaneous control over the structure in the in-plane directions is so far still rather limited. Here, we combine spatially controlled modifications of 2D materials, using focused electron irradiation or electron beam induced etching, with the layer-by-layer assembly of van der Waals heterostructures. A novel assembly process makes it possible to structure each layer with an arbitrary pattern prior to the assembly into the heterostructure. Moreover, it enables a stacking of the layers with accurate lateral alignment, with an accuracy of currently 10nm, under observation in an electron microscope. Together, this enables the fabrication of almost arbitrary 3D structures with highest spatial resolution.

physics.app-ph↗

arXiv2021

Magnetic Frustration in a Zeolite

Zeolites are so well known in real world applications and after decades of scientific study that they hardly need any intro-duction: their importance in chemistry cannot be overemphasized. Here we add to the remarkable properties that they dis-play by reporting our discovery that the simplest zeolite, sodalite, when doped with Cr3+ in the \b{eta}-cage, is a frustrated magnet. Soft X-ray absorption spectroscopy and magnetic measurements reveal that the Cr present is Cr(III). Cr(III), with its isotropic 3d3 valence electron configuration, is well-known as the basis for many geometrically frustrated magnets, but it is especially surprising that a material like the Ca8Al12Cr2O29 zeolite is a frustrated magnet. This finding illustrates the value of exploring the properties of even well-known materials families.

cond-mat.mtrl-sci↗

arXiv2021

Secure Machine Learning over Relational Data

A closer integration of machine learning and relational databases has gained steam in recent years due to the fact that the training data to many ML tasks is the results of a relational query (most often, a join-select query). In a federated setting, this poses an additional challenge, that the tables are held by different parties as their private data, and the parties would like to train the model without having to use a trusted third party. Existing work has only considered the case where the training data is stored in a flat table that has been vertically partitioned, which corresponds to a simple PK-PK join. In this paper, we describe secure protocols to compute the join results of multiple tables conforming to a general foreign-key acyclic schema, and how to feed the results in secret-shared form to a secure ML toolbox. Furthermore, existing secure ML systems reveal the PKs in the join results. We strengthen the privacy protection to higher levels and achieve zero information leakage beyond the trained model. If the model itself is considered sensitive, we show how differential privacy can be incorporated into our framework to also prevent the model from breaching individuals' privacy.

cs.CR↗

arXiv2021

CoSeg: Cognitively Inspired Unsupervised Generic Event Segmentation

Some cognitive research has discovered that humans accomplish event segmentation as a side effect of event anticipation. Inspired by this discovery, we propose a simple yet effective end-to-end self-supervised learning framework for event segmentation/boundary detection. Unlike the mainstream clustering-based methods, our framework exploits a transformer-based feature reconstruction scheme to detect event boundary by reconstruction errors. This is consistent with the fact that humans spot new events by leveraging the deviation between their prediction and what is actually perceived. Thanks to their heterogeneity in semantics, the frames at boundaries are difficult to be reconstructed (generally with large reconstruction errors), which is favorable for event boundary detection. Additionally, since the reconstruction occurs on the semantic feature level instead of pixel level, we develop a temporal contrastive feature embedding module to learn the semantic visual representation for frame feature reconstruction. This procedure is like humans building up experiences with "long-term memory". The goal of our work is to segment generic events rather than localize some specific ones. We focus on achieving accurate event boundaries. As a result, we adopt F1 score (Precision/Recall) as our primary evaluation metric for a fair comparison with previous approaches. Meanwhile, we also calculate the conventional frame-based MoF and IoU metric. We thoroughly benchmark our work on four publicly available datasets and demonstrate much better results.

cs.CV↗

arXiv2021

Inferential Wasserstein Generative Adversarial Networks

Generative Adversarial Networks (GANs) have been impactful on many problems and applications but suffer from unstable training. The Wasserstein GAN (WGAN) leverages the Wasserstein distance to avoid the caveats in the minmax two-player training of GANs but has other defects such as mode collapse and lack of metric to detect the convergence. We introduce a novel inferential Wasserstein GAN (iWGAN) model, which is a principled framework to fuse auto-encoders and WGANs. The iWGAN model jointly learns an encoder network and a generator network motivated by the iterative primal dual optimization process. The encoder network maps the observed samples to the latent space and the generator network maps the samples from the latent space to the data space. We establish the generalization error bound of the iWGAN to theoretically justify its performance. We further provide a rigorous probabilistic interpretation of our model under the framework of maximum likelihood estimation. The iWGAN, with a clear stopping criteria, has many advantages over other autoencoder GANs. The empirical experiments show that the iWGAN greatly mitigates the symptom of mode collapse, speeds up the convergence, and is able to provide a measurement of quality check for each individual sample. We illustrate the ability of the iWGAN by obtaining competitive and stable performances for benchmark datasets.

stat.ML↗

arXiv2021

Constants of Motion: The Antidote to Chaos in Optimization and Game Dynamics

Several recent works in online optimization and game dynamics have established strong negative complexity results including the formal emergence of instability and chaos even in small such settings, e.g., $2\times 2$ games. These results motivate the following question: Which methodological tools can guarantee the regularity of such dynamics and how can we apply them in standard settings of interest such as discrete-time first-order optimization dynamics? We show how proving the existence of invariant functions, i.e., constant of motions, is a fundamental contribution in this direction and establish a plethora of such positive results (e.g. gradient descent, multiplicative weights update, alternating gradient descent and manifold gradient descent) both in optimization as well as in game settings. At a technical level, for some conservation laws we provide an explicit and concise closed form, whereas for other ones we present non-constructive proofs using tools from dynamical systems.

math.OC↗

arXiv2021

STADB: A Self-Thresholding Attention Guided ADB Network for Person Re-identification

Recently, Batch DropBlock network (BDB) has demonstrated its effectiveness on person image representation and re-identification task via feature erasing. However, BDB drops the features \textbf{randomly} which may lead to sub-optimal results. In this paper, we propose a novel Self-Thresholding attention guided Adaptive DropBlock network (STADB) for person re-ID which can \textbf{adaptively} erase the most discriminative regions. Specifically, STADB first obtains an attention map by channel-wise pooling and returns a drop mask by thresholding the attention map. Then, the input features and self-thresholding attention guided drop mask are multiplied to generate the dropped feature maps. In addition, STADB utilizes the spatial and channel attention to learn a better feature map and iteratively trains the feature dropping module for person re-ID. Experiments on several benchmark datasets demonstrate that the proposed STADB outperforms many other related methods for person re-ID. The source code of this paper is released at: \textcolor{red}{\url{https://github.com/wangxiao5791509/STADB_ReID}}.

cs.CV↗

arXiv2021

Learn to Match: Automatic Matching Network Design for Visual Tracking

Siamese tracking has achieved groundbreaking performance in recent years, where the essence is the efficient matching operator cross-correlation and its variants. Besides the remarkable success, it is important to note that the heuristic matching network design relies heavily on expert experience. Moreover, we experimentally find that one sole matching operator is difficult to guarantee stable tracking in all challenging environments. Thus, in this work, we introduce six novel matching operators from the perspective of feature fusion instead of explicit similarity learning, namely Concatenation, Pointwise-Addition, Pairwise-Relation, FiLM, Simple-Transformer and Transductive-Guidance, to explore more feasibility on matching operator selection. The analyses reveal these operators' selective adaptability on different environment degradation types, which inspires us to combine them to explore complementary features. To this end, we propose binary channel manipulation (BCM) to search for the optimal combination of these operators. BCM determines to retrain or discard one operator by learning its contribution to other tracking steps. By inserting the learned matching networks to a strong baseline tracker Ocean, our model achieves favorable gains by $67.2 \rightarrow 71.4$, $52.6 \rightarrow 58.3$, $70.3 \rightarrow 76.0$ success on OTB100, LaSOT, and TrackingNet, respectively. Notably, Our tracker, dubbed AutoMatch, uses less than half of training data/time than the baseline tracker, and runs at 50 FPS using PyTorch. Code and model will be released at https://github.com/JudasDie/SOTS.

cs.CV↗

arXiv2021

Semantic-guided Pixel Sampling for Cloth-Changing Person Re-identification

Cloth-changing person re-identification (re-ID) is a new rising research topic that aims at retrieving pedestrians whose clothes are changed. This task is quite challenging and has not been fully studied to date. Current works mainly focus on body shape or contour sketch, but they are not robust enough due to view and posture variations. The key to this task is to exploit cloth-irrelevant cues. This paper proposes a semantic-guided pixel sampling approach for the cloth-changing person re-ID task. We do not explicitly define which feature to extract but force the model to automatically learn cloth-irrelevant cues. Specifically, we first recognize the pedestrian's upper clothes and pants, then randomly change them by sampling pixels from other pedestrians. The changed samples retain the identity labels but exchange the pixels of clothes or pants among different pedestrians. Besides, we adopt a loss function to constrain the learned features to keep consistent before and after changes. In this way, the model is forced to learn cues that are irrelevant to upper clothes and pants. We conduct extensive experiments on the latest released PRCC dataset. Our method achieved 65.8% on Rank1 accuracy, which outperforms previous methods with a large margin. The code is available at https://github.com/shuxjweb/pixel_sampling.git.

cs.CV↗

arXiv2021

Robust Pipek-Mezey Orbital Localization in Periodic Solids

We describe a robust method for determining Pipek-Mezey (PM) Wannier functions (WF), recently introduced by Jónsson et al. (J. Chem. Theor. Chem. 2017, 13, 460), which provide some formal advantages over the more common Boys (also known as maximally-localized) Wannier functions. The Broyden-Fletcher-Goldfarb-Shanno (BFGS) based PMWF solver is demonstrated to yield dramatically faster convergence compared to the alternatives (steepest ascent and conjugate gradient) in a variety of 1-, 2-, and 3-dimensional solids (including some with vanishing gaps), and can be used to obtain Wannier functions robustly in supercells with thousands of atoms. Evaluation of the PM functional and its gradient in periodic LCAO representation used a particularly simple definition of atomic charges obtained by Moore-Penrose pseudoinverse projection onto the minimal atomic orbital basis. An automated "Canonicalize Phase then Randomize" (CPR) method for generating the initial guess for WFs contributes significantly to the robustness of the solver.

physics.chem-ph↗

arXiv2021

Absorption Spectra of Solids from Periodic Equation-of-Motion Coupled-Cluster Theory

We present ab initio absorption spectra of six three-dimensional semiconductors and insulators calculated using Gaussian-based periodic equation-of-motion coupled-cluster theory with single and double excitations (EOM-CCSD). The spectra are calculated efficiently by solving a system of linear equations at each frequency, giving access to an energy range of tens of eV without explicit enumeration of excited states. We assess the impact of Brillouin zone sampling, for which it is hard to achieve convergence due to the cost of EOM-CCSD. Although our most converged spectra exhibit lineshapes that are in good agreement with experiment, they are uniformly shifted to higher energies by about 1 eV. We tentatively attribute this discrepancy to a combination of vibrational effects and the remaining electron correlation, i.e., triple excitations and above.

physics.chem-ph↗

arXiv2021

Topological magnon insulators in two-dimensional van der Waals ferromagnets CrSiTe$_3$ and CrGeTe$_3$: towards intrinsic gap-tunability

The bosonic analogues of topological insulators have been proposed in numerous theoretical works, but their experimental realization is still very rare, especially for spin systems. Recently, two-dimensional (2D) honeycomb van der Waals (vdW) ferromagnets have emerged as a new platform for topological spin excitations. Here, via a comprehensive inelastic neutron scattering study and theoretical analysis of the spin-wave excitations, we report the realization of topological magnon insulators in CrXTe$_3$ (X=Si, Ge) compounds. The nontrivial nature and intrinsic tunability of the gap opening at the magnon band-crossing Dirac points are confirmed, while the emergence of the corresponding in-gap topological edge states is demonstrated theoretically. The realization of topological magnon insulators with intrinsic gap-tunability in this class of remarkable 2D materials will undoubtedly lead to new and fascinating technological applications in the domain of magnonics and topological spintronics.

cond-mat.str-el↗

arXiv2021

Gaze Estimation with an Ensemble of Four Architectures

This paper presents a method for gaze estimation according to face images. We train several gaze estimators adopting four different network architectures, including an architecture designed for gaze estimation (i.e.,iTracker-MHSA) and three originally designed for general computer vision tasks(i.e., BoTNet, HRNet, ResNeSt). Then, we select the best six estimators and ensemble their predictions through a linear combination. The method ranks the first on the leader-board of ETH-XGaze Competition, achieving an average angular error of $3.11^{\circ}$ on the ETH-XGaze test set.

cs.CV↗

arXiv2021

Pseudo-Relevance Feedback for Multiple Representation Dense Retrieval

Pseudo-relevance feedback mechanisms, from Rocchio to the relevance models, have shown the usefulness of expanding and reweighting the users' initial queries using information occurring in an initial set of retrieved documents, known as the pseudo-relevant set. Recently, dense retrieval -- through the use of neural contextual language models such as BERT for analysing the documents' and queries' contents and computing their relevance scores -- has shown a promising performance on several information retrieval tasks still relying on the traditional inverted index for identifying documents relevant to a query. Two different dense retrieval families have emerged: the use of single embedded representations for each passage and query (e.g. using BERT's [CLS] token), or via multiple representations (e.g. using an embedding for each token of the query and document). In this work, we conduct the first study into the potential for multiple representation dense retrieval to be enhanced using pseudo-relevance feedback. In particular, based on the pseudo-relevant set of documents identified using a first-pass dense retrieval, we extract representative feedback embeddings (using KMeans clustering) -- while ensuring that these embeddings discriminate among passages (based on IDF) -- which are then added to the query representation. These additional feedback embeddings are shown to both enhance the effectiveness of a reranking as well as an additional dense retrieval operation. Indeed, experiments on the MSMARCO passage ranking dataset show that MAP can be improved by upto 26% on the TREC 2019 query set and 10% on the TREC 2020 query set by the application of our proposed ColBERT-PRF method on a ColBERT dense retrieval approach.

cs.IR↗

arXiv2021

Unraveling intrinsic flexoelectricity in twisted double bilayer graphene

Moiré superlattices of two-dimensional (2D) materials with a small twist angle are thought to exhibit appreciable flexoelectric effect, though unambiguous confirmation of their flexoelectricity is challenging due to artifacts associated with commonly used piezoresponse force microscopy (PFM). For example, unexpectedly small phase contrast ($\sim$$8^{\circ}$) between opposite flexoelectric polarizations was reported in twisted bilayer graphene (tBG), though theoretically predicted value is $180^{\circ}$. Here we developed a methodology to extract intrinsic moiré flexoelectricity using twisted double bilayer graphene (tDBG) as a model system, probed by lateral PFM. For small twist angle samples, we found that a vectorial decomposition is essential to recover the small intrinsic flexoelectric response at domain walls from a large background signal. The obtained three-fold symmetry of commensurate domains with significant flexoelectric response at domain walls is fully consistent with our theoretical calculations. Incommensurate domains in tDBG with relatively large twist angles can also be observed by this technique. Our work provides a general strategy for unraveling intrinsic flexoelectricity in van der Waals moiré superlattices while providing insights into engineered symmetry breaking in centrosymmetric materials.

cond-mat.mtrl-sci↗

arXiv2021

Atomic-scale imaging of CH3NH3PbI3 structure and its decomposition pathway

Understanding the atomic structure and structural instability of organic-inorganic hybrid perovskites is the key to appreciate their remarkable photoelectric properties and failure mechanism. Here, using low-dose imaging technique by direct-detection electron-counting camera in transmission electron microscope, we investigate the atomic structure and decomposition pathway of CH3NH3PbI3 (MAPbI3) at the atomic scale. We successfully image the atomic structure of perovskite in real space under ultra-low electron dose condition, and observe a two-step decomposition process, i.e. initial loss of MA followed by the collapse of perovskite structure into 6H-PbI2 with their critical threshold dose also determined. Interestingly, an intermediate phase (MA0.5PbI3) with locally ordered vacancies can robustly exist before perovskite collapses, enlightening strategies for prevention and recovery of perovskite structure during degradation. Associated with structure evolution, the bandgap gradually increases from ~1.6 eV to ~2.1 eV, and it is found that both C-N and N-H bonds can be destroyed under irradiation, releasing NH3 and leaving hydrocarbons. These findings enhance our understanding of the photoelectric properties and failure mechanism of MAPbI3, providing potential strategy into material optimization.

physics.app-ph↗

arXiv2021

Cascade of singularities in the spin dynamics of a perturbed quantum critical Ising chain

When the quantum critical transverse-field Ising chain is perturbed by a longitudinal field, a quantum integrable model emerges in the scaling limit with massive excitations described by the exceptional $E_{8}$ Lie algebra. Using the corresponding analytical form factors of the quantum $E_{8}$ integrable model, we systematically study the spin dynamic structure factor of the perturbed quantum critical Ising chain, where particle channels with total energy up to 5$m_1$ ($m_1$ being the mass of the lightest $E_{8}$ particle) are exhausted. In addition to the significant single-particle contributions to the dynamic spectrum, each two-particle channel with different masses is found to exhibit an edge singularity at the threshold of the total mass and decays with an inverse square root of energy, which is attributed to the singularity of the two-particle density of states at the threshold. The singularity is absent for particles with equal masses due to a cancellation mechanism involving the structure of the form factors. As a consequence, the dynamic structure factor displays a cascade of bumping peaks in the continuum region with clear singular features which can serve as a solid guidance for the material realization of the quantum $E_{8}$ model.

cond-mat.str-el↗

arXiv2021

Tracking by Joint Local and Global Search: A Target-aware Attention based Approach

Tracking-by-detection is a very popular framework for single object tracking which attempts to search the target object within a local search window for each frame. Although such local search mechanism works well on simple videos, however, it makes the trackers sensitive to extremely challenging scenarios, such as heavy occlusion and fast motion. In this paper, we propose a novel and general target-aware attention mechanism (termed TANet) and integrate it with tracking-by-detection framework to conduct joint local and global search for robust tracking. Specifically, we extract the features of target object patch and continuous video frames, then we concatenate and feed them into a decoder network to generate target-aware global attention maps. More importantly, we resort to adversarial training for better attention prediction. The appearance and motion discriminator networks are designed to ensure its consistency in spatial and temporal views. In the tracking procedure, we integrate the target-aware attention with multiple trackers by exploring candidate search regions for robust tracking. Extensive experiments on both short-term and long-term tracking benchmark datasets all validated the effectiveness of our algorithm. The project page of this paper can be found at \url{https://sites.google.com/view/globalattentiontracking/home/extend}.

cs.CV↗