arXiv subjects

Zhao Zhang

Publications and source records attributed to Zhao Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2022

OSIC: A New One-Stage Image Captioner Coined

Mainstream image caption models are usually two-stage captioners, i.e., calculating object features by pre-trained detector, and feeding them into a language model to generate text descriptions. However, such an operation will cause a task-based information gap to decrease the performance, since the object features in detection task are suboptimal representation and cannot provide all necessary information for subsequent text generation. Besides, object features are usually represented by the last layer features that lose the local details of input images. In this paper, we propose a novel One-Stage Image Captioner (OSIC) with dynamic multi-sight learning, which directly transforms input image into descriptive sentences in one stage. As a result, the task-based information gap can be greatly reduced. To obtain rich features, we use the Swin Transformer to calculate multi-level features, and then feed them into a novel dynamic multi-sight embedding module to exploit both global structure and local texture of input images. To enhance the global modeling of encoder for caption, we propose a new dual-dimensional refining module to non-locally model the interaction of the embedded features. Finally, OSIC can obtain rich and useful information to improve the image caption task. Extensive comparisons on benchmark MS-COCO dataset verified the superior performance of our method.

cs.CV↗

arXiv2022

Unifying Graph Embedding Features with Graph Convolutional Networks for Skeleton-based Action Recognition

Combining skeleton structure with graph convolutional networks has achieved remarkable performance in human action recognition. Since current research focuses on designing basic graph for representing skeleton data, these embedding features contain basic topological information, which cannot learn more systematic perspectives from skeleton data. In this paper, we overcome this limitation by proposing a novel framework, which unifies 15 graph embedding features into the graph convolutional network for human action recognition, aiming to best take advantage of graph information to distinguish key joints, bones, and body parts in human action, instead of being exclusive to a single feature or domain. Additionally, we fully investigate how to find the best graph features of skeleton structure for improving human action recognition. Besides, the topological information of the skeleton sequence is explored to further enhance the performance in a multi-stream framework. Moreover, the unified graph features are extracted by the adaptive methods on the training process, which further yields improvements. Our model is validated by three large-scale datasets, namely NTU-RGB+D, Kinetics and SYSU-3D, and outperforms the state-of-the-art methods. Overall, our work unified graph embedding features to promotes systematic research on human action recognition.

cs.CV↗

arXiv2022

Evolution is Still Good: Theoretical Analysis of Evolutionary Algorithms on General Cover Problems

Theoretical studies on evolutionary algorithms have developed vigorously in recent years. Many such algorithms have theoretical guarantees in both running time and approximation ratio. Some approximation mechanism seems to be inherently embedded in many evolutionary algorithms. In this paper, we identify such a relation by proposing a unified analysis framework for a generalized simple multi-objective evolutionary algorithm (GSEMO), and apply it on a minimum weight general cover problem. For a wide range of problems (including the the minimum submodular cover problem in which the submodular function is real-valued, and the minimum connected dominating set problem for which the potential function is non-submodular), GSEMO yields asymptotically tight approximation ratios in expected polynomial time.

cs.NE↗

arXiv2022

A Simple Approach to Automated Spectral Clustering

The performance of spectral clustering heavily relies on the quality of affinity matrix. A variety of affinity-matrix-construction (AMC) methods have been proposed but they have hyperparameters to determine beforehand, which requires strong experience and leads to difficulty in real applications, especially when the inter-cluster similarity is high and/or the dataset is large. In addition, we often need to choose different AMC methods for different datasets, which still depends on experience. To solve these two challenging problems, in this paper, we present a simple yet effective method for automated spectral clustering. First, we propose to find the most reliable affinity matrix via grid search or Bayesian optimization among a set of candidates given by different AMC methods with different hyperparameters, where the reliability is quantified by the \textit{relative-eigen-gap} of graph Laplacian introduced in this paper. Second, we propose a fast and accurate AMC method based on least squares representation and thresholding and prove its effectiveness theoretically. Finally, we provide a large-scale extension for the automated spectral clustering method, of which the time complexity is linear with the number of data points. Extensive experiments of natural image clustering show that our method is more versatile, accurate, and efficient than baseline methods.

cs.LG↗

arXiv2022

ASTF: Visual Abstractions of Time-Varying Patterns in Radio Signals

A time-frequency diagram is a commonly used visualization for observing the time-frequency distribution of radio signals and analyzing their time-varying patterns of communication states in radio monitoring and management. While it excels when performing short-term signal analyses, it becomes inadaptable for long-term signal analyses because it cannot adequately depict signal time-varying patterns in a large time span on a space-limited screen. This research thus presents an abstract signal time-frequency (ASTF) diagram to address this problem. In the diagram design, a visual abstraction method is proposed to visually encode signal communication state changes in time slices. A time segmentation algorithm is proposed to divide a large time span into time slices.Three new quantified metrics and a loss function are defined to ensure the preservation of important time-varying information in the time segmentation. An algorithm performance experiment and a user study are conducted to evaluate the effectiveness of the diagram for long-term signal analyses.

eess.SP↗

arXiv2022

PointScatter: Point Set Representation for Tubular Structure Extraction

This paper explores the point set representation for tubular structure extraction tasks. Compared with the traditional mask representation, the point set representation enjoys its flexibility and representation ability, which would not be restricted by the fixed grid as the mask. Inspired by this, we propose PointScatter, an alternative to the segmentation models for the tubular structure extraction task. PointScatter splits the image into scatter regions and parallelly predicts points for each scatter region. We further propose the greedy-based region-wise bipartite matching algorithm to train the network end-to-end and efficiently. We benchmark the PointScatter on four public tubular datasets, and the extensive experiments on tubular structure segmentation and centerline extraction task demonstrate the effectiveness of our approach. Code is available at https://github.com/zhangzhao2022/pointscatter.

cs.CV↗

arXiv2022

Glassy crystals with colossal multi-baroresponsivities

As a nontrivial solid state of matter, the glassy-crystal state embraces physical features of both crystalline and amorphous solids, where a long-range ordered periodic structure formed by the mass centers of constituent molecules accommodates orientational glasses. Here, we discover and validate a glassy-crystal state in 2-amino-2-methyl-1,3-propanediol (AMP, C4H11NO2) by neutron scattering and complementary broadband dielectric spectroscopy (BDS) measurements. The freezing process of the dynamic orientational disorder is manifested at relaxation times well described by the Vogel-Fulcher-Tammann (VFT) law and the strongly frequency-dependent freezing temperature ranging from around 225 K at 0.1 Hz to above room temperature in the GHz region. At room temperature, the supercooled state is extremely sensitive to pressure such that a few MPa pressure can induce crystallization to the ordered crystal state, eventually leading to a temperature increase by 48 K within 20 s, a significant reduction of visible light transmittance from about 95% to a few percentages, and a remarkable decrease of electrical conductivity by three orders of magnitude. These ultrasensitive baroresponsivities might find their applications in low-grade waste heat recycling, pressure sensors and non-volatile memory devices. It is expected that glassy crystals serve as an emerging platform for exploiting exotic states of matter and the associated fantastic applications.

cond-mat.mtrl-sci↗

arXiv2022

A Multi-Task Learning Model for Super Resolution of Wireless Channel Characteristics

Channel modeling has always been the core part in communication system design and development, especially in 5G and 6G era. Traditional approaches like stochastic channel modeling and ray-tracing (RT) based channel modeling depend heavily on measurement data or simulation, which are usually expensive and time consuming. In this paper, we propose a novel super resolution (SR) model for generating channel characteristics data. The model is based on multi-task learning (MTL) convolutional neural networks (CNN) with residual connection. Experiments demonstrate that the proposed SR model could achieve excellent performances in mean absolute error and standard deviation of error. Advantages of the proposed model are demonstrated in comparisons with other state-of-the-art deep learning models. Ablation study also proved the necessity of multi-task learning and techniques in model design. The contribution in this paper could be helpful in channel modeling, network optimization, positioning and other wireless channel characteristics related work by largely reducing workload of simulation or measurement.

eess.SP↗

arXiv2022

Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting

We all depend on mobility, and vehicular transportation affects the daily lives of most of us. Thus, the ability to forecast the state of traffic in a road network is an important functionality and a challenging task. Traffic data is often obtained from sensors deployed in a road network. Recent proposals on spatial-temporal graph neural networks have achieved great progress at modeling complex spatial-temporal correlations in traffic data, by modeling traffic data as a diffusion process. However, intuitively, traffic data encompasses two different kinds of hidden time series signals, namely the diffusion signals and inherent signals. Unfortunately, nearly all previous works coarsely consider traffic signals entirely as the outcome of the diffusion, while neglecting the inherent signals, which impacts model performance negatively. To improve modeling performance, we propose a novel Decoupled Spatial-Temporal Framework (DSTF) that separates the diffusion and inherent traffic information in a data-driven manner, which encompasses a unique estimation gate and a residual decomposition mechanism. The separated signals can be handled subsequently by the diffusion and inherent modules separately. Further, we propose an instantiation of DSTF, Decoupled Dynamic Spatial-Temporal Graph Neural Network (D2STGNN), that captures spatial-temporal correlations and also features a dynamic graph learning module that targets the learning of the dynamic characteristics of traffic networks. Extensive experiments with four real-world traffic datasets demonstrate that the framework is capable of advancing the state-of-the-art.

cs.LG↗

arXiv2022

Hidden SU(2)_D vector dark matter with a scalar septuplet

We propose a vector dark matter model from a hidden SU(2)_D gauge symmetry at TeV scale. A scalar septuplet is introduced to break the SU(2)_D symmetry spontaneously. The septuplet also play the role of a portal between the standard model and the dark sector. We find that there are two different vacuum configurations corresponding to the sign of the quartic coupling $λ_3$, which yields different mass spectrum for the gauge bosons. For a $λ_3<0$, the masses of gauge bosons are splitting, while for a $λ_3\geq0$, the masses are degenerate. We also study the RG evolutions of the couplings, and find that the perturbativity and vacuum stability can set a stringent bound on the parameter space. For the phenomenological aspect, we consider the experimental constraints including dark matter direct detection, indirect detection, relic density, and Higgs couplings measurements. We find that there are parameter space survive from all the constraints, and they can be tested in future dark matter direct and indirect detection experiments.

hep-ph↗

arXiv2022

Approximation Algorithm for Minimum $p$ Union Under a Geometric Setting

In a minimum $p$ union problem (Min$p$U), given a hypergraph $G=(V,E)$ and an integer $p$, the goal is to find a set of $p$ hyperedges $E'\subseteq E$ such that the number of vertices covered by $E'$ (that is $|\bigcup_{e\in E'}e|$) is minimized. It was known that Min$p$U is at least as hard as the densest $k$-subgraph problem. A question is: how about the problem in some geometric settings? In this paper, we consider the unit square Min$p$U problem (Min$p$U-US) in which $V$ is a set of points on the plane, and each hyperedge of $E$ consists of a set of points in a unit square. A $(\frac{1}{1+\varepsilon},4)$-bicriteria approximation algorithm is presented, that is, the algorithm finds at least $\frac{p}{1+\varepsilon}$ unit squares covering at most $4opt$ points, where $opt$ is the optimal value for the Min$p$U-US instance (the minimum number of points that can be covered by $p$ unit squares).

cs.CG↗

arXiv2022

A Hierarchical Interactive Network for Joint Span-based Aspect-Sentiment Analysis

Recently, some span-based methods have achieved encouraging performances for joint aspect-sentiment analysis, which first extract aspects (aspect extraction) by detecting aspect boundaries and then classify the span-level sentiments (sentiment classification). However, most existing approaches either sequentially extract task-specific features, leading to insufficient feature interactions, or they encode aspect features and sentiment features in a parallel manner, implying that feature representation in each task is largely independent of each other except for input sharing. Both of them ignore the internal correlations between the aspect extraction and sentiment classification. To solve this problem, we novelly propose a hierarchical interactive network (HI-ASA) to model two-way interactions between two tasks appropriately, where the hierarchical interactions involve two steps: shallow-level interaction and deep-level interaction. First, we utilize cross-stitch mechanism to combine the different task-specific features selectively as the input to ensure proper two-way interactions. Second, the mutual information technique is applied to mutually constrain learning between two tasks in the output layer, thus the aspect input and the sentiment input are capable of encoding features of the other task via backpropagation. Extensive experiments on three real-world datasets demonstrate HI-ASA's superiority over baselines.

cs.CL↗

arXiv2022

Hidden Bethe states in a partially integrable model

We present a one-dimensional multi-component model, known to be partially integrable when restricted to the subspaces made of only two components. By constructing fully anti-symmetrized bases, we find integrable excited eigenstates corresponding to the totally anti-symmetric irreducible representation of the permutation operator in the otherwise non-integrable subspaces. We establish rigorously the breakdown of integrability in those subspaces by showing explicitly the violation of the Yang-Baxter's equation. We further solve the constraints from Yang-Baxter's equation to find exceptional momenta that allows Bethe Ansatz solutions of solitonic bound states. These integrable eigenstates have distinct dynamical consequence from the embedded integrable subspaces previously known, as they do not span their separate Krylov subspaces, and a generic initial state can partly overlap with them and therefore have slow thermalization. However, this novel form of weak ergodicity breaking contrasts that of quantum many-body scars in that the integrable eigenstates involved do not have necessarily low entanglement. Our approach provides a complementary route to arrive at quantum many-body scars since, instead of solving towers of single mode excited states based on a solvable ground state in a non-integrable model, we identify the integrable eigenstates that survive in a deformation of the Hamiltonian away from its integrable point.

cond-mat.stat-mech↗

arXiv2022

Spatial-Temporal Identity: A Simple yet Effective Baseline for Multivariate Time Series Forecasting

Multivariate Time Series (MTS) forecasting plays a vital role in a wide range of applications. Recently, Spatial-Temporal Graph Neural Networks (STGNNs) have become increasingly popular MTS forecasting methods due to their state-of-the-art performance. However, recent works are becoming more sophisticated with limited performance improvements. This phenomenon motivates us to explore the critical factors of MTS forecasting and design a model that is as powerful as STGNNs, but more concise and efficient. In this paper, we identify the indistinguishability of samples in both spatial and temporal dimensions as a key bottleneck, and propose a simple yet effective baseline for MTS forecasting by attaching Spatial and Temporal IDentity information (STID), which achieves the best performance and efficiency simultaneously based on simple Multi-Layer Perceptrons (MLPs). These results suggest that we can design efficient and effective models as long as they solve the indistinguishability of samples, without being limited to STGNNs.

cs.LG↗

arXiv2022

A Survey on Incomplete Multi-view Clustering

Conventional multi-view clustering seeks to partition data into respective groups based on the assumption that all views are fully observed. However, in practical applications, such as disease diagnosis, multimedia analysis, and recommendation system, it is common to observe that not all views of samples are available in many cases, which leads to the failure of the conventional multi-view clustering methods. Clustering on such incomplete multi-view data is referred to as incomplete multi-view clustering. In view of the promising application prospects, the research of incomplete multi-view clustering has noticeable advances in recent years. However, there is no survey to summarize the current progresses and point out the future research directions. To this end, we review the recent studies of incomplete multi-view clustering. Importantly, we provide some frameworks to unify the corresponding incomplete multi-view clustering methods, and make an in-depth comparative analysis for some representative methods from theoretical and experimental perspectives. Finally, some open problems in the incomplete multi-view clustering field are offered for researchers.

cs.LG↗

arXiv2022

Pre-training Enhanced Spatial-temporal Graph Neural Network for Multivariate Time Series Forecasting

Multivariate Time Series (MTS) forecasting plays a vital role in a wide range of applications. Recently, Spatial-Temporal Graph Neural Networks (STGNNs) have become increasingly popular MTS forecasting methods. STGNNs jointly model the spatial and temporal patterns of MTS through graph neural networks and sequential models, significantly improving the prediction accuracy. But limited by model complexity, most STGNNs only consider short-term historical MTS data, such as data over the past one hour. However, the patterns of time series and the dependencies between them (i.e., the temporal and spatial patterns) need to be analyzed based on long-term historical MTS data. To address this issue, we propose a novel framework, in which STGNN is Enhanced by a scalable time series Pre-training model (STEP). Specifically, we design a pre-training model to efficiently learn temporal patterns from very long-term history time series (e.g., the past two weeks) and generate segment-level representations. These representations provide contextual information for short-term time series input to STGNNs and facilitate modeling dependencies between time series. Experiments on three public real-world datasets demonstrate that our framework is capable of significantly enhancing downstream STGNNs, and our pre-training model aptly captures temporal patterns.

cs.LG↗

arXiv2022

FCL-GAN: A Lightweight and Real-Time Baseline for Unsupervised Blind Image Deblurring

Blind image deblurring (BID) remains a challenging and significant task. Benefiting from the strong fitting ability of deep learning, paired data-driven supervised BID method has obtained great progress. However, paired data are usually synthesized by hand, and the realistic blurs are more complex than synthetic ones, which makes the supervised methods inept at modeling realistic blurs and hinders their real-world applications. As such, unsupervised deep BID method without paired data offers certain advantages, but current methods still suffer from some drawbacks, e.g., bulky model size, long inference time, and strict image resolution and domain requirements. In this paper, we propose a lightweight and real-time unsupervised BID baseline, termed Frequency-domain Contrastive Loss Constrained Lightweight CycleGAN (shortly, FCL-GAN), with attractive properties, i.e., no image domain limitation, no image resolution limitation, 25x lighter than SOTA, and 5x faster than SOTA. To guarantee the lightweight property and performance superiority, two new collaboration units called lightweight domain conversion unit(LDCU) and parameter-free frequency-domain contrastive unit(PFCU) are designed. LDCU mainly implements inter-domain conversion in lightweight manner. PFCU further explores the similarity measure, external difference and internal connection between the blurred domain and sharp domain images in frequency domain, without involving extra parameters. Extensive experiments on several image datasets demonstrate the effectiveness of our FCL-GAN in terms of performance, model size and reference time.

cs.CV↗

arXiv2022

Customized Conversational Recommender Systems

Conversational recommender systems (CRS) aim to capture user's current intentions and provide recommendations through real-time multi-turn conversational interactions. As a human-machine interactive system, it is essential for CRS to improve the user experience. However, most CRS methods neglect the importance of user experience. In this paper, we propose two key points for CRS to improve the user experience: (1) Speaking like a human, human can speak with different styles according to the current dialogue context. (2) Identifying fine-grained intentions, even for the same utterance, different users have diverse finegrained intentions, which are related to users' inherent preference. Based on the observations, we propose a novel CRS model, coined Customized Conversational Recommender System (CCRS), which customizes CRS model for users from three perspectives. For human-like dialogue services, we propose multi-style dialogue response generator which selects context-aware speaking style for utterance generation. To provide personalized recommendations, we extract user's current fine-grained intentions from dialogue context with the guidance of user's inherent preferences. Finally, to customize the model parameters for each user, we train the model from the meta-learning perspective. Extensive experiments and a series of analyses have shown the superiority of our CCRS on both the recommendation and dialogue services.

cs.IR↗