arXiv subjects

Jie Wang

Publications and source records attributed to Jie Wang.

At least 433 records · Page 24Linked to original sources

arXiv2022

Tag-Set-Sequence Learning for Generating Question-Answer Pairs

Transformer-based QG models can generate question-answer pairs (QAPs) with high qualities, but may also generate silly questions for certain texts. We present a new method called tag-set sequence learning to tackle this problem, where a tag-set sequence is a sequence of tag sets to capture the syntactic and semantic information of the underlying sentence, and a tag set consists of one or more language feature tags, including, for example, semantic-role-labeling, part-of-speech, named-entity-recognition, and sentiment-indication tags. We construct a system called TSS-Learner to learn tag-set sequences from given declarative sentences and the corresponding interrogative sentences, and derive answers to the latter. We train a TSS-Learner model for the English language using a small training dataset and show that it can indeed generate adequate QAPs for certain texts that transformer-based models do poorly. Human evaluation on the QAPs generated by TSS-Learner over SAT practice reading tests is encouraging.

cs.CL↗

arXiv2022

Incommensurate Magnetic Order in Hole-Doped Infinite-layer Nickelate Superconductors

Magnetism and superconductivity are closely entangled, elucidating the magnetic interactions in nickelate superconductors is at the heart of understanding the pairing mechanism. Our first-principles and spin-wave theory calculations highlight that NdNiO$_2$ is in the vicinity of a transition between a quasi-two-dimensional (2D) antiferromagnetic (AFM) state and a three-dimensional (3D) C-AFM state. Both states could accurately reproduce the experimentally measured magnetic excitation spectra, which was previously explained in terms of a 2D model. We further reveal that hole doping stabilizes an incommensurate (IC) spin state and the IC wave vector increases continuously. Direct links between hole doping, magnetization, exchange constants, and magnetic order are established, revealing that the competition between first-neighbor and third-neighbor in-plane magnetic interactions is the key for the IC magnetic order.

cond-mat.supr-con↗

arXiv2022

Spatial-aware Speaker Diarization for Multi-channel Multi-party Meeting

This paper describes a spatial-aware speaker diarization system for the multi-channel multi-party meeting. The diarization system obtains direction information of speaker by microphone array. Speaker spatial embedding is generated by xvector and s-vector derived from superdirective beamforming (SDB) which makes the embedding more robust. Specifically, we propose a novel multi-channel sequence-to-sequence neural network architecture named discriminative multi-stream neural network (DMSNet) which consists of attention superdirective beamforming (ASDB) block and Conformer encoder. The proposed ASDB is a self-adapted channel-wise block that extracts the latent spatial features of array audios by modeling interdependencies between channels. We explore DMSNet to address overlapped speech problem on multi-channel audio and achieve 93.53% accuracy on evaluation set. By performing DMSNet based overlapped speech detection (OSD) module, the diarization error rate (DER) of cluster-based diarization system decrease significantly from 13.45% to 7.64%.

eess.AS↗

arXiv2022

FusionRCNN: LiDAR-Camera Fusion for Two-stage 3D Object Detection

3D object detection with multi-sensors is essential for an accurate and reliable perception system of autonomous driving and robotics. Existing 3D detectors significantly improve the accuracy by adopting a two-stage paradigm which merely relies on LiDAR point clouds for 3D proposal refinement. Though impressive, the sparsity of point clouds, especially for the points far away, making it difficult for the LiDAR-only refinement module to accurately recognize and locate objects.To address this problem, we propose a novel multi-modality two-stage approach named FusionRCNN, which effectively and efficiently fuses point clouds and camera images in the Regions of Interest(RoI). FusionRCNN adaptively integrates both sparse geometry information from LiDAR and dense texture information from camera in a unified attention mechanism. Specifically, it first utilizes RoIPooling to obtain an image set with a unified size and gets the point set by sampling raw points within proposals in the RoI extraction step; then leverages an intra-modality self-attention to enhance the domain-specific features, following by a well-designed cross-attention to fuse the information from two modalities.FusionRCNN is fundamentally plug-and-play and supports different one-stage methods with almost no architectural changes. Extensive experiments on KITTI and Waymo benchmarks demonstrate that our method significantly boosts the performances of popular detectors.Remarkably, FusionRCNN significantly improves the strong SECOND baseline by 6.14% mAP on Waymo, and outperforms competing two-stage approaches. Code will be released soon at https://github.com/xxlbigbrother/Fusion-RCNN.

cs.CV↗

arXiv2022

Democratizing Domain-Specific Computing

In the past few years, domain-specific accelerators (DSAs), such as Google's Tensor Processing Units, have shown to offer significant performance and energy efficiency over general-purpose CPUs. An important question is whether typical software developers can design and implement their own customized DSAs, with affordability and efficiency, to accelerate their applications. This article presents our answer to this question.

cs.AR↗

arXiv2022

TAPA: A Scalable Task-Parallel Dataflow Programming Framework for Modern FPGAs with Co-Optimization of HLS and Physical Design

In this paper, we propose TAPA, an end-to-end framework that compiles a C++ task-parallel dataflow program into a high-frequency FPGA accelerator. Compared to existing solutions, TAPA has two major advantages. First, TAPA provides a set of convenient APIs that allow users to easily express flexible and complex inter-task communication structures. Second, TAPA adopts a coarse-grained floorplanning step during HLS compilation for accurate pipelining of potential critical paths. In addition, TAPA implements several optimization techniques specifically tailored for modern HBM-based FPGAs. In our experiments with a total of 43 designs, we improve the average frequency from 147 MHz to 297 MHz (a 102% improvement) with no loss of throughput and a negligible change in resource utilization. Notably, in 16 experiments we make the originally unroutable designs achieve 274 MHz on average. The framework is available at https://github.com/UCLA-VAST/tapa and the core floorplan module is available at https://github.com/UCLA-VAST/AutoBridge.

cs.AR↗

arXiv2022

Contrastive Semantic-Guided Image Smoothing Network

Image smoothing is a fundamental low-level vision task that aims to preserve salient structures of an image while removing insignificant details. Deep learning has been explored in image smoothing to deal with the complex entanglement of semantic structures and trivial details. However, current methods neglect two important facts in smoothing: 1) naive pixel-level regression supervised by the limited number of high-quality smoothing ground-truth could lead to domain shift and cause generalization problems towards real-world images; 2) texture appearance is closely related to object semantics, so that image smoothing requires awareness of semantic difference to apply adaptive smoothing strengths. To address these issues, we propose a novel Contrastive Semantic-Guided Image Smoothing Network (CSGIS-Net) that combines both contrastive prior and semantic prior to facilitate robust image smoothing. The supervision signal is augmented by leveraging undesired smoothing effects as negative teachers, and by incorporating segmentation tasks to encourage semantic distinctiveness. To realize the proposed network, we also enrich the original VOC dataset with texture enhancement and smoothing labels, namely VOC-smooth, which first bridges image smoothing and semantic segmentation. Extensive experiments demonstrate that the proposed CSGIS-Net outperforms state-of-the-art algorithms by a large margin. Code and dataset are available at https://github.com/wangjie6866/CSGIS-Net.

cs.CV↗

arXiv2022

Stability of Linear Systems under Extended Weakly-Hard Constraints

Control systems can show robustness to many events, like disturbances and model inaccuracies. It is natural to speculate that they are also robust to sporadic deadline misses when implemented as digital tasks on an embedded platform. This paper proposes a comprehensive stability analysis for control systems subject to deadline misses, leveraging a new formulation to describe the patterns experienced by the control task under different handling strategies. Such analysis brings the assessment of control systems robustness to computational problems one step closer to the controller implementation.

math.OC↗

arXiv2022

Sparse Polynomial Optimization: Theory and Practice

The problem of minimizing a polynomial over a set of polynomial inequalities is an NP-hard non-convex problem. Thanks to powerful results from real algebraic geometry, one can convert this problem into a nested sequence of finite-dimensional convex problems. At each step of the associated hierarchy, one needs to solve a fixed size semidefinite program, which can be in turn solved with efficient numerical tools. On the practical side however, there is no-free lunch and such optimization methods usually encompass severe scalability issues. Fortunately, for many applications, we can look at the problem in the eyes and exploit the inherent data structure arising from the cost and constraints describing the problem, for instance sparsity or symmetries. This book presents several research efforts to tackle this scientific challenge with important computational implications, and provides the development of alternative optimization schemes that scale well in terms of computational complexity, at least in some identified class of problems. The presented algorithmic framework in this book mainly exploits the sparsity structure of the input data to solve large-scale polynomial optimization problems. We present sparsity-exploiting hierarchies of relaxations, for either unconstrained or constrained problems. By contrast with the dense hierarchies, they provide faster approximation of the solution in practice but also come with the same theoretical convergence guarantees. Our framework is not restricted to static polynomial optimization, and we expose hierarchies of approximations for values of interest arising from the analysis of dynamical systems. We also present various extensions to problems involving noncommuting variables, e.g., matrices of arbitrary size or quantum physic operators.

math.OC↗

arXiv2022

Dirac electron under periodic magnetic field: Platform for fractional Chern insulator and generalized Wigner crystal

We propose a platform for flat Chern band by subjecting two-dimensional Dirac materials -- such as graphene and topological insulator thin films -- to a periodic magnetic field, which can be created by the vortex lattice of a type-II superconductor. As a generalization of the $n=0$ Landau level, the flat band of Dirac fermion under a nonuniform magnetic field remains at zero energy, exactly dispersionless and topologically protected, while its local density of states is spatially modulated due to the magnetic field variation. In the presence of short-range repulsion, we find fractional Chern insulators emerge at filling factors $ν=1/m$, whose ground states are generalized Laughlin wavefunctions. We further argue that generalized Wigner crystals may emerge at certain commensurate fillings under a highly nonuniform magnetic field in the form of a flux line lattice.

cond-mat.mes-hall↗

arXiv2022

Speech Representation Disentanglement with Adversarial Mutual Information Learning for One-shot Voice Conversion

One-shot voice conversion (VC) with only a single target speaker's speech for reference has become a hot research topic. Existing works generally disentangle timbre, while information about pitch, rhythm and content is still mixed together. To perform one-shot VC effectively with further disentangling these speech components, we employ random resampling for pitch and content encoder and use the variational contrastive log-ratio upper bound of mutual information and gradient reversal layer based adversarial mutual information learning to ensure the different parts of the latent space containing only the desired disentangled representation during training. Experiments on the VCTK dataset show the model achieves state-of-the-art performance for one-shot VC in terms of naturalness and intellgibility. In addition, we can transfer characteristics of one-shot VC on timbre, pitch and rhythm separately by speech representation disentanglement. Our code, pre-trained models and demo are available at https://im1eon.github.io/IS2022-SRDVC/.

eess.AS↗

arXiv2022

Modeling Diverse Chemical Reactions for Single-step Retrosynthesis via Discrete Latent Variables

Single-step retrosynthesis is the cornerstone of retrosynthesis planning, which is a crucial task for computer-aided drug discovery. The goal of single-step retrosynthesis is to identify the possible reactants that lead to the synthesis of the target product in one reaction. By representing organic molecules as canonical strings, existing sequence-based retrosynthetic methods treat the product-to-reactant retrosynthesis as a sequence-to-sequence translation problem. However, most of them struggle to identify diverse chemical reactions for a desired product due to the deterministic inference, which contradicts the fact that many compounds can be synthesized through various reaction types with different sets of reactants. In this work, we aim to increase reaction diversity and generate various reactants using discrete latent variables. We propose a novel sequence-based approach, namely RetroDVCAE, which incorporates conditional variational autoencoders into single-step retrosynthesis and associates discrete latent variables with the generation process. Specifically, RetroDVCAE uses the Gumbel-Softmax distribution to approximate the categorical distribution over potential reactions and generates multiple sets of reactants with the variational decoder. Experiments demonstrate that RetroDVCAE outperforms state-of-the-art baselines on both benchmark dataset and homemade dataset. Both quantitative and qualitative results show that RetroDVCAE can model the multi-modal distribution over reaction types and produce diverse reactant candidates.

q-bio.QM↗

arXiv2022

An Efficient End-to-End 3D Voxel Reconstruction based on Neural Architecture Search

Using neural networks to represent 3D objects has become popular. However, many previous works employ neural networks with fixed architecture and size to represent different 3D objects, which lead to excessive network parameters for simple objects and limited reconstruction accuracy for complex objects. For each 3D model, it is desirable to have an end-to-end neural network with as few parameters as possible to achieve high-fidelity reconstruction. In this paper, we propose an efficient voxel reconstruction method utilizing neural architecture search (NAS) and binary classification. Taking the number of layers, the number of nodes in each layer, and the activation function of each layer as the search space, a specific network architecture can be obtained based on reinforcement learning technology. Furthermore, to get rid of the traditional surface reconstruction algorithms (e.g., marching cube) used after network inference, we complete the end-to-end network by classifying binary voxels. Compared to other signed distance field (SDF) prediction or binary classification networks, our method achieves significantly higher reconstruction accuracy using fewer network parameters.

cs.CV↗

arXiv2022

Improving Pseudo Labels With Intra-Class Similarity for Unsupervised Domain Adaptation

Unsupervised domain adaptation (UDA) transfers knowledge from a label-rich source domain to a different but related fully-unlabeled target domain. To address the problem of domain shift, more and more UDA methods adopt pseudo labels of the target samples to improve the generalization ability on the target domain. However, inaccurate pseudo labels of the target samples may yield suboptimal performance with error accumulation during the optimization process. Moreover, once the pseudo labels are generated, how to remedy the generated pseudo labels is far from explored. In this paper, we propose a novel approach to improve the accuracy of the pseudo labels in the target domain. It first generates coarse pseudo labels by a conventional UDA method. Then, it iteratively exploits the intra-class similarity of the target samples for improving the generated coarse pseudo labels, and aligns the source and target domains with the improved pseudo labels. The accuracy improvement of the pseudo labels is made by first deleting dissimilar samples, and then using spanning trees to eliminate the samples with the wrong pseudo labels in the intra-class samples. We have applied the proposed approach to several conventional UDA methods as an additional term. Experimental results demonstrate that the proposed method can boost the accuracy of the pseudo labels and further lead to more discriminative and domain invariant features than the conventional baselines.

cs.CV↗

arXiv2022

Certifying Global Optimality of AC-OPF Solutions via sparse polynomial optimization

We report the experimental results on certifying 1% global optimality of solutions of AC-OPF instances from PGLiB via the CS-TSSOS hierarchy -- a moment-SOS based hierarchy that exploits both correlative and term sparsity, which can provide tighter SDP relaxations than Shor's relaxation. Our numerical experiments demonstrate that the CS-TSSOS hierarchy scales well with the problem size and is indeed useful in certifying global optimality of solutions for large-scale real world problems, e.g., the AC-OPF problem. In particular, we are able to certify 1% global optimality for a challenging AC-OPF instance with 6515 buses involving 14398 real variables and 63577 constraints.

math.OC↗

arXiv2022

Milky Way Mass with K Giants and BHB Stars Using LAMOST, SDSS/SEGUE, and Gaia: 3D Spherical Jeans Equation and Tracer Mass Estimator

We measure the enclosed Milky Way mass profile to Galactocentric distances of $\sim70$ and $\sim50$ kpc using the smooth, diffuse stellar halo samples of Bird et al. The samples are LAMOST and SDSS/SEGUE K giants (KG) and SDSS/SEGUE blue horizontal branch (BHB) stars with accurate metallicities. The 3D kinematics are available through LAMOST and SDSS/SEGUE distances and radial velocities and {\it Gaia} DR2 proper motions. Two methods are used to estimate the enclosed mass: 3D spherical Jeans equation and Evans et al. tracer mass estimator (TME). We remove substructure via the Xue et al. method based on integrals of motion. We evaluate the uncertainties on our estimates due to random sampling noise, systematic distance errors, the adopted density profile, and non-virialization and non-spherical effects of the halo. The tracer density profile remains a limiting systematic in our mass estimates, although within these limits we find reasonable agreement across the different samples and the methods applied. Out to $\sim70$ and $\sim50$ kpc, the Jeans method yields total enclosed masses of $4.3\pm0.95$ (random) $\pm0.6$ (systematic) $\times10^{11}$ M$_\odot$ and $4.1\pm1.2$ (random) $\pm0.6$ (systematic) $\times10^{11}$ M$_\odot$ for the KG and BHB stars, respectively. For the KG and BHB samples we find a dark matter virial mass of $M_{200}=0.55^{+0.15}_{-0.11}$ (random) $\pm0.083$ (systematic) $\times10^{12}$ M$_\odot$ and $M_{200}=1.00^{+0.67}_{-0.33}$ (random) $\pm0.15$ (systematic) $\times10^{12}$ M$_\odot$, respectively.

astro-ph.GA↗

arXiv2022

TMGAN-PLC: Audio Packet Loss Concealment using Temporal Memory Generative Adversarial Network

Real-time communications in packet-switched networks have become widely used in daily communication, while they inevitably suffer from network delays and data losses in constrained real-time conditions. To solve these problems, audio packet loss concealment (PLC) algorithms have been developed to mitigate voice transmission failures by reconstructing the lost information. Limited by the transmission latency and device memory, it is still intractable for PLC to accomplish high-quality voice reconstruction using a relatively small packet buffer. In this paper, we propose a temporal memory generative adversarial network for audio PLC, dubbed TMGAN-PLC, which is comprised of a novel nested-UNet generator and the time-domain/frequency-domain discriminators. Specifically, a combination of the nested-UNet and temporal feature-wise linear modulation is elaborately devised in the generator to finely adjust the intra-frame information and establish inter-frame temporal dependencies. To complement the missing speech content caused by longer loss bursts, we employ multi-stage gated vector quantizers to capture the correct content and reconstruct the near-real smooth audio. Extensive experiments on the PLC Challenge dataset demonstrate that the proposed method yields promising performance in terms of speech quality, intelligibility, and PLCMOS.

cs.SD↗

arXiv2022

About One-point Statistics of the Ratio of Two Fourier-transformed Cosmic Fields and an Application

The Fourier transformation is an effective and efficient operation of Gaussianization at the one-point level. Using a set of N-body simulation data, we verified that the one-point distribution functions of the dark matter momentum divergence and density fields closely follow complex Gaussian distributions. The one-point distribution function of the quotient of two complex Gaussian variables is introduced and studied. Statistical theories are then applied to model one-point statistics about the growth of individual Fourier mode of the dark matter density field, which can be obtained by the ratio of two Fourier transformed cosmic fields. Our simulation results proved that the models based on the Gaussian approximation are impressively accurate, and our analysis revealed many interesting aspects about the growth of dark matter's density fluctuation in Fourier space.

astro-ph.CO↗