arXiv subjects

Yu Wang

Publications and source records attributed to Yu Wang.

At least 1,027 records · Page 57Linked to original sources

arXiv2022

Latent Heterogeneous Graph Network for Incomplete Multi-View Learning

Multi-view learning has progressed rapidly in recent years. Although many previous studies assume that each instance appears in all views, it is common in real-world applications for instances to be missing from some views, resulting in incomplete multi-view data. To tackle this problem, we propose a novel Latent Heterogeneous Graph Network (LHGN) for incomplete multi-view learning, which aims to use multiple incomplete views as fully as possible in a flexible manner. By learning a unified latent representation, a trade-off between consistency and complementarity among different views is implicitly realized. To explore the complex relationship between samples and latent representations, a neighborhood constraint and a view-existence constraint are proposed, for the first time, to construct a heterogeneous graph. Finally, to avoid any inconsistencies between training and test phase, a transductive learning technique is applied based on graph learning for classification tasks. Extensive experimental results on real-world datasets demonstrate the effectiveness of our model over existing state-of-the-art approaches.

cs.LG↗

arXiv2022

Hyperradial distribution function of few-body problems: a new arena for extreme value theory

This work explores classical capture models for few-body systems via a Monte Carlo method in hyperspherical coordinates. In particular, we focus on van der Waals and charged-induced dipole interactions. As a result, we notice that, independently of the number of particles and interparticle interaction, the capture hyperradial distribution function follows a Fréchet distribution, a special type of the generalized extreme value distribution. Besides, we elaborate on the fundamentals of such universal feature using the general extreme value theory, thus, establishing a connection between extreme value theory and few-body physics.

physics.chem-ph↗

arXiv2022

BSDGAN: Balancing Sensor Data Generative Adversarial Networks for Human Activity Recognition

The development of IoT technology enables a variety of sensors can be integrated into mobile devices. Human Activity Recognition (HAR) based on sensor data has become an active research topic in the field of machine learning and ubiquitous computing. However, due to the inconsistent frequency of human activities, the amount of data for each activity in the human activity dataset is imbalanced. Considering the limited sensor resources and the high cost of manually labeled sensor data, human activity recognition is facing the challenge of highly imbalanced activity datasets. In this paper, we propose Balancing Sensor Data Generative Adversarial Networks (BSDGAN) to generate sensor data for minority human activities. The proposed BSDGAN consists of a generator model and a discriminator model. Considering the extreme imbalance of human activity dataset, an autoencoder is employed to initialize the training process of BSDGAN, ensure the data features of each activity can be learned. The generated activity data is combined with the original dataset to balance the amount of activity data across human activity classes. We deployed multiple human activity recognition models on two publicly available imbalanced human activity datasets, WISDM and UNIMIB. Experimental results show that the proposed BSDGAN can effectively capture the data features of real human activity sensor data, and generate realistic synthetic sensor data. Meanwhile, the balanced activity dataset can effectively help the activity recognition model to improve the recognition accuracy.

cs.LG↗

arXiv2022

An electron-spin qubit platform assembled atom-by-atom on a surface

Creating a quantum-coherent architecture at the atomic scale has long been an ambition in quantum science and nanotechnology. This ultimate length scale requires the use of fundamental quantum properties of atoms, such as the spin of electrons, which naturally occurs in many solid-state environments and allows high-fidelity operations and readout by electromagnetic means. Despite decades of effort, however, it remains a formidable task to realize an atomic-scale quantum architecture where multiple electron spin qubits can be precisely assembled, controllably coupled, and coherently operated. Electron spin qubits created in dopants in semiconductors and color centers in insulators, for example, can be well controlled individually6-8 but are difficult to couple together into a circuit. On the other hand, multiple magnetic atoms and molecules on surfaces can be coupled to each other by building sophisticated atomic structures using a scanning tunneling microscope (STM), but coherent operation has so far been limited to a single qubit in the tunnel junction. Here we demonstrate an atomic-scale qubit platform by showing atom-by-atom construction, coherent operations, and readout of multiple electron-spin qubits on a surface. To enable the coherent control of remote qubits that are outside the tunnel junction, we complement each electron spin with a local magnetic field gradient from a nearby single-atom magnet. To enable readout of remote qubits, we employ a sensor qubit in the tunnel junction and implement pulsed double electron spin resonance. Using these methods, we demonstrate fast single-, two-, and three-qubit operations in an all-electrical fashion. Our work marks the creation of an Angstrom-scale qubit platform, where quantum functionalities using electron spin arrays, built atom-by-atom on a surface, are now within reach.

cond-mat.mes-hall↗

arXiv2022

Chiral SO(4) spin-valley density wave and degenerate topological superconductivity in magic-angle-twisted bilayer-graphene

Starting from a realistic extended Hubbard model for a $p_{x,y}$-orbital tight-binding model on the Honeycomb lattice, we perform a thorough investigation on the possible electron instabilities in the MA-TBG near the van Hove (VH) dopings. Here we focus on the interplay between the approximate SU(2)$\times$SU(2) symmetry and the $D_3$ symmetry, which leads to intriguing quantum states relevant to recent experiments, as revealed by our systematic RPA based calculations followed by a succeeding mean-field energy minimization for the ground state energy. At the SU(2)$\times$SU(2) symmetric point, the degenerate inter-valley SDW and VDW are mixed into a new state of matter dubbed as the chiral SO(4) spin-valley DW. This state simultaneously hosts three 4-component vectorial spin-valley DW orders with each adopting one wave vector, and the polarization directions of the three DW orders are mutually perpendicular to one another. %in the $\mathbb{R}^4$ space. In the presence of a tiny inter-valley exchange interaction with coefficient $J_H\to 0^{-}$ which breaks the SU(2)$\times$SU(2) symmetry, a pure chiral SDW state is obtained. In the case of $J_H\to 0^{+}$, a nematic VDW+SDW state emerges which possesses a stripy distribution of the charge density, consistent with the recent STM observations. On the aspect of SC, while the triplet $p+ip$ and singlet $d+id$ topological SCs are degenerate at $J_H=0$ near the VH dopings, the former (latter) is favored for $J_H\to 0^{-}$ ($J_H\to 0^{+}$). In addition, the two asymmetric doping-dependent behaviors of the obtained pairing phase diagram are well consistent with experiments.

cond-mat.str-el↗

arXiv2022

Strong structural and electronic coupling in metavalent PbS moire superlattices

Moire superlattices are twisted bilayer materials, in which the tunable interlayer quantum confinement offers access to new physics and novel device functionalities. Previously, moire superlattices were built exclusively using materials with weak van der Waals interactions and synthesizing moire superlattices with strong interlayer chemical bonding was considered to be impractical. Here using lead sulfide (PbS) as an example, we report a strategy for synthesizing of moire superlattices coupled by strong chemical bonding. We use water-soluble ligands as a removable template to obtain free-standing ultra-thin PbS nanosheets and assemble them into direct-contact bilayers with various twist angles. Atomic-resolution imaging shows the moire periodic structural reconstruction at superlattice interface, due to the strong metavalent coupling. Electron energy loss spectroscopy and theoretical calculations collectively reveal the twist angle26 dependent electronic structure, especially the emergent separation of flat bands at small twist angles. The localized states of flat bands are similar to well-arranged quantum dots, promising an application in devices. This study opens a new door to the exploration of deep energy modulations within moire superlattices alternative to van der Waals twistronics.

cond-mat.mtrl-sci↗

arXiv2022

CLOSE: Curriculum Learning On the Sharing Extent Towards Better One-shot NAS

One-shot Neural Architecture Search (NAS) has been widely used to discover architectures due to its efficiency. However, previous studies reveal that one-shot performance estimations of architectures might not be well correlated with their performances in stand-alone training because of the excessive sharing of operation parameters (i.e., large sharing extent) between architectures. Thus, recent methods construct even more over-parameterized supernets to reduce the sharing extent. But these improved methods introduce a large number of extra parameters and thus cause an undesirable trade-off between the training costs and the ranking quality. To alleviate the above issues, we propose to apply Curriculum Learning On Sharing Extent (CLOSE) to train the supernet both efficiently and effectively. Specifically, we train the supernet with a large sharing extent (an easier curriculum) at the beginning and gradually decrease the sharing extent of the supernet (a harder curriculum). To support this training strategy, we design a novel supernet (CLOSENet) that decouples the parameters from operations to realize a flexible sharing scheme and adjustable sharing extent. Extensive experiments demonstrate that CLOSE can obtain a better ranking quality across different computational budget constraints than other one-shot supernets, and is able to discover superior architectures when combined with various search strategies. Code is available at https://github.com/walkerning/aw_nas.

cs.CV↗

arXiv2022

Few-Shot Specific Emitter Identification via Deep Metric Ensemble Learning

Specific emitter identification (SEI) is a highly potential technology for physical layer authentication that is one of the most critical supplement for the upper-layer authentication. SEI is based on radio frequency (RF) features from circuit difference, rather than cryptography. These features are inherent characteristic of hardware circuits, which difficult to counterfeit. Recently, various deep learning (DL)-based conventional SEI methods have been proposed, and achieved advanced performances. However, these methods are proposed for close-set scenarios with massive RF signal samples for training, and they generally have poor performance under the condition of limited training samples. Thus, we focus on few-shot SEI (FS-SEI) for aircraft identification via automatic dependent surveillance-broadcast (ADS-B) signals, and a novel FS-SEI method is proposed, based on deep metric ensemble learning (DMEL). Specifically, the proposed method consists of feature embedding and classification. The former is based on metric learning with complex-valued convolutional neural network (CVCNN) for extracting discriminative features with compact intra-category distance and separable inter-category distance, while the latter is realized by an ensemble classifier. Simulation results show that if the number of samples per category is more than 5, the average accuracy of our proposed method is higher than 98\%. Moreover, feature visualization demonstrates the advantages of our proposed method in both discriminability and generalization. The codes of this paper can be downloaded from GitHub(https://github.com/BeechburgPieStar/Few-Shot-Specific-Emitter-Identification-via-Deep-Metric-Ensemble-Learning)

eess.SP↗

arXiv2022

WheaCha: A Method for Explaining the Predictions of Models of Code

Attribution methods have emerged as a popular approach to interpreting model predictions based on the relevance of input features. Although the feature importance ranking can provide insights of how models arrive at a prediction from a raw input, they do not give a clear-cut definition of the key features models use for the prediction. In this paper, we present a new method, called WheaCha, for explaining the predictions of code models. Although WheaCha employs the same mechanism of tracing model predictions back to the input features, it differs from all existing attribution methods in crucial ways. Specifically, WheaCha divides an input program into "wheat" (i.e., the defining features that are the reason for which models predict the label that they predict) and the rest "chaff" for any prediction of a learned code model. We realize WheaCha in a tool, HuoYan, and use it to explain four prominent code models: code2vec, seq-GNN, GGNN, and CodeBERT. Results show (1) HuoYan is efficient - taking on average under twenty seconds to compute the wheat for an input program in an end-to-end fashion (i.e., including model prediction time); (2) the wheat that all models use to predict input programs is made of simple syntactic or even lexical properties (i.e., identifier names); (3) Based on wheat, we present a novel approach to explaining the predictions of code models through the lens of training data.

cs.LG↗

arXiv2022

Multi-level Fusion of Wav2vec 2.0 and BERT for Multimodal Emotion Recognition

The research and applications of multimodal emotion recognition have become increasingly popular recently. However, multimodal emotion recognition faces the challenge of lack of data. To solve this problem, we propose to use transfer learning which leverages state-of-the-art pre-trained models including wav2vec 2.0 and BERT for this task. Multi-level fusion approaches including coattention-based early fusion and late fusion with the models trained on both embeddings are explored. Also, a multi-granularity framework which extracts not only frame-level speech embeddings but also segment-level embeddings including phone, syllable and word-level speech embeddings is proposed to further boost the performance. By combining our coattention-based early fusion model and late fusion model with the multi-granularity feature extraction framework, we obtain result that outperforms best baseline approaches by 1.3% unweighted accuracy (UA) on the IEMOCAP dataset.

cs.CL↗

arXiv2022

Dual Vision Transformer

Prior works have proposed several strategies to reduce the computational cost of self-attention mechanism. Many of these works consider decomposing the self-attention procedure into regional and local feature extraction procedures that each incurs a much smaller computational complexity. However, regional information is typically only achieved at the expense of undesirable information lost owing to down-sampling. In this paper, we propose a novel Transformer architecture that aims to mitigate the cost issue, named Dual Vision Transformer (Dual-ViT). The new architecture incorporates a critical semantic pathway that can more efficiently compress token vectors into global semantics with reduced order of complexity. Such compressed global semantics then serve as useful prior information in learning finer pixel level details, through another constructed pixel pathway. The semantic pathway and pixel pathway are then integrated together and are jointly trained, spreading the enhanced self-attention information in parallel through both of the pathways. Dual-ViT is henceforth able to reduce the computational complexity without compromising much accuracy. We empirically demonstrate that Dual-ViT provides superior accuracy than SOTA Transformer architectures with reduced training complexity. Source code is available at \url{https://github.com/YehLi/ImageNetModel}.

cs.CV↗

arXiv2022

GRB 190829A -- A Showcase of Binary Late Evolution

GRB 190829A is the fourth closest gamma-ray burst (GRB) to date ($z=0.0785$). Owing to its wide range of radio, optical, X-ray, and the very-high-energy (VHE) observations by H.E.S.S., it has become an essential new source examined by various models with complementary approaches. We here show in GRB 190829A the double-prompt pulses and the three-multiwavelength afterglows are consistent with the type II binary-driven hypernova (BdHN II) model. The progenitor is a binary composed of a carbon-oxygen (CO) star and a neutron star (NS) companion. The gravitational collapse of the iron core of the CO star produces a supernova (SN) explosion and leaves behind a new neutron star ($ν$NS) at its center. The accretion of the SN ejecta onto the NS companion and onto the $ν$NS via matter fallback spins up the NSs and produces the double-peak prompt emission. The synchrotron emission from the expanding SN ejecta with the energy injection from the rapidly spinning $ν$NS and its subsequently spin-down leads to the afterglow in the radio, optical, and X-ray. We model the sequence of physical and related radiation processes in BdHNe and focus on individuating the binary properties that play the relevant roles.

astro-ph.HE↗

arXiv2022

Class-Specific Semantic Reconstruction for Open Set Recognition

Open set recognition enables deep neural networks (DNNs) to identify samples of unknown classes, while maintaining high classification accuracy on samples of known classes. Existing methods basing on auto-encoder (AE) and prototype learning show great potential in handling this challenging task. In this study, we propose a novel method, called Class-Specific Semantic Reconstruction (CSSR), that integrates the power of AE and prototype learning. Specifically, CSSR replaces prototype points with manifolds represented by class-specific AEs. Unlike conventional prototype-based methods, CSSR models each known class on an individual AE manifold, and measures class belongingness through AE's reconstruction error. Class-specific AEs are plugged into the top of the DNN backbone and reconstruct the semantic representations learned by the DNN instead of the raw image. Through end-to-end learning, the DNN and the AEs boost each other to learn both discriminative and representative information. The results of experiments conducted on multiple datasets show that the proposed method achieves outstanding performance in both close and open set recognition and is sufficiently simple and flexible to incorporate into existing frameworks.

cs.CV↗

arXiv2022

Multi-Granularity Regularized Re-Balancing for Class Incremental Learning

Deep learning models suffer from catastrophic forgetting when learning new tasks incrementally. Incremental learning has been proposed to retain the knowledge of old classes while learning to identify new classes. A typical approach is to use a few exemplars to avoid forgetting old knowledge. In such a scenario, data imbalance between old and new classes is a key issue that leads to performance degradation of the model. Several strategies have been designed to rectify the bias towards the new classes due to data imbalance. However, they heavily rely on the assumptions of the bias relation between old and new classes. Therefore, they are not suitable for complex real-world applications. In this study, we propose an assumption-agnostic method, Multi-Granularity Regularized re-Balancing (MGRB), to address this problem. Re-balancing methods are used to alleviate the influence of data imbalance; however, we empirically discover that they would under-fit new classes. To this end, we further design a novel multi-granularity regularization term that enables the model to consider the correlations of classes in addition to re-balancing the data. A class hierarchy is first constructed by grouping the semantically or visually similar classes. The multi-granularity regularization then transforms the one-hot label vector into a continuous label distribution, which reflects the relations between the target class and other classes based on the constructed class hierarchy. Thus, the model can learn the inter-class relational information, which helps enhance the learning of both old and new classes. Experimental results on both public datasets and a real-world fault diagnosis dataset verify the effectiveness of the proposed method.

cs.CV↗

arXiv2022

Differentially Private Algorithms for Statistical Verification of Cyber-Physical Systems

Statistical model checking is a class of sequential algorithms that can verify specifications of interest on an ensemble of cyber-physical systems (e.g., whether 99% of cars from a batch meet a requirement on their energy efficiency). These algorithms infer the probability that given specifications are satisfied by the systems with provable statistical guarantees by drawing sufficient numbers of independent and identically distributed samples. During the process of statistical model checking, the values of the samples (e.g., a user's car energy efficiency) may be inferred by intruders, causing privacy concerns in consumer-level applications (e.g., automobiles and medical devices). This paper addresses the privacy of statistical model checking algorithms from the point of view of differential privacy. These algorithms are sequential, drawing samples until a condition on their values is met. We show that revealing the number of the samples drawn can violate privacy. We also show that the standard exponential mechanism that randomizes the output of an algorithm to achieve differential privacy fails to do so in the context of sequential algorithms. Instead, we relax the conservative requirement in differential privacy that the sensitivity of the output of the algorithm should be bounded to any perturbation for any data set. We propose a new notion of differential privacy which we call expected differential privacy. Then, we propose a novel expected sensitivity analysis for the sequential algorithm and proposed a corresponding exponential mechanism that randomizes the termination time to achieve the expected differential privacy. We apply the proposed mechanism to statistical model checking algorithms to preserve the privacy of the samples they draw. The utility of the proposed algorithm is demonstrated in a case study.

cs.LG↗

arXiv2022

LidarMultiNet: Unifying LiDAR Semantic Segmentation, 3D Object Detection, and Panoptic Segmentation in a Single Multi-task Network

This technical report presents the 1st place winning solution for the Waymo Open Dataset 3D semantic segmentation challenge 2022. Our network, termed LidarMultiNet, unifies the major LiDAR perception tasks such as 3D semantic segmentation, object detection, and panoptic segmentation in a single framework. At the core of LidarMultiNet is a strong 3D voxel-based encoder-decoder network with a novel Global Context Pooling (GCP) module extracting global contextual features from a LiDAR frame to complement its local features. An optional second stage is proposed to refine the first-stage segmentation or generate accurate panoptic segmentation results. Our solution achieves a mIoU of 71.13 and is the best for most of the 22 classes on the Waymo 3D semantic segmentation test set, outperforming all the other 3D semantic segmentation methods on the official leaderboard. We demonstrate for the first time that major LiDAR perception tasks can be unified in a single strong network that can be trained end-to-end.

cs.CV↗

arXiv2022

On Structural Explanation of Bias in Graph Neural Networks

Graph Neural Networks (GNNs) have shown satisfying performance in various graph analytical problems. Hence, they have become the \emph{de facto} solution in a variety of decision-making scenarios. However, GNNs could yield biased results against certain demographic subgroups. Some recent works have empirically shown that the biased structure of the input network is a significant source of bias for GNNs. Nevertheless, no studies have systematically scrutinized which part of the input network structure leads to biased predictions for any given node. The low transparency on how the structure of the input network influences the bias in GNN outcome largely limits the safe adoption of GNNs in various decision-critical scenarios. In this paper, we study a novel research problem of structural explanation of bias in GNNs. Specifically, we propose a novel post-hoc explanation framework to identify two edge sets that can maximally account for the exhibited bias and maximally contribute to the fairness level of the GNN prediction for any given node, respectively. Such explanations not only provide a comprehensive understanding of bias/fairness of GNN predictions but also have practical significance in building an effective yet fair GNN model. Extensive experiments on real-world datasets validate the effectiveness of the proposed framework towards delivering effective structural explanations for the bias of GNNs. Open-source code can be found at https://github.com/yushundong/REFEREE.

cs.LG↗

arXiv2022

Pure State Tomography with Fourier Transformation

Extracting information from quantum devices has long been a crucial problem in the field of quantum mechanics. By performing elaborate measurements, quantum state tomography, an important and fundamental tool in quantum science and technology, can be used to determine unknown quantum states completely. In this study, we explore methods to determine multi-qubit pure quantum states uniquely and directly. Two adaptive protocols are proposed, with their respective quantum circuits. Herein, two or three observables are sufficient, while the number of measurement outcomes is either the same as or fewer than those in existing methods. Additionally, experiments on the IBM 5-qubit quantum computer, as well as numerical investigations, demonstrate the feasibility of the proposed protocols.

quant-ph↗