Arrow Research search

Author name cluster

Xiang Wu

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

15 papers
1 author row

Possible papers

15

EAAI Journal 2026 Journal Article

Attention dynamic graph convolutional network for traffic flow prediction

  • Chenhui Wei
  • Chuanming Chen
  • Xiang Wu
  • Dongmei Pan
  • Qingying Yu
  • Xiaoyao Zheng
  • YongLong Luo

Traffic flow prediction is crucial for intelligent transportation systems, enabling effective urban planning, traffic management, and emergency response. Existing methods rely on adjacency matrices to model traffic network connections, often failing to capture real-time dynamics and complex spatiotemporal dependencies due to static or data-limited dynamic matrices. To address these challenges, we propose an Attention Dynamic Graph Convolutional Network (ADGCN) that integrates an adaptive dynamic graph convolutional network with a novel lightweight Gated Recurrent Unit and an enhanced attention mechanism. The lightweight gated recurrent unit offers significant efficiency gains over the standard gated recurrent unit. By optimizing its gating mechanism and shrinking the hidden layer dimension, it features a lower number of parameters and achieves a 19 %–47 % reduction in training time. These improvements make it highly suitable for deployment on resource-constrained devices and for use in real-time traffic applications. The dynamic graph generation method, leveraging input features and node embeddings with normalization and nonlinear transformations, effectively captures evolving spatial dependencies without predefined graph structures, enhancing adaptability to fluctuating traffic conditions. The improved attention mechanism strengthens inter-channel feature dependencies, enabling the model to focus on task-critical features and boosting prediction accuracy. Validated on six real-world datasets, ADGCN outperforms most state-of-the-art models on key metrics. It also demonstrates remarkable efficiency, with up to a 47 % reduction in training time and excellent real-time performance, making it highly suitable for Intelligent Transportation Systems.

Details DOI

EAAI Journal 2026 Journal Article

Integration of multiple constraint-treating approaches for uncertain automatic parking path optimization utilizing competition and cooperation-driven differential evolution

  • Xiang Wu

To better alleviate driver stress and guarantee safe parking, this paper re-tackles an optimal autonomous vehicle (AV) parking operation problem with uncertainty. A probability constrained dynamic optimization problem (PCDOP) is constructed for the parking operation to obtain a path that moves AV from a pre-specified location to a parking position while minimizing time and avoiding collisions. Directly handling PCDOP is challenging due to the difficulty of deriving explicit formulas for the probability functions in probability path-constraints. To tackle this, a tractable deterministic constrained parameter selection problem (DCPSP) is developed to approximate it utilizing average sampling, discretization, and time-scaling. The constraints in DCPSP are usually multi-layered, dynamic, and conflicting. Thus, a single constraint-treating approach (CTA) is insufficient to address such complex constraints. For this reason, a competition and cooperation-driven differential evolution integrating multiple constraint-treating approaches (CCDDE-IMCTAs) is designed for DCPSP, utilizing a competition-driven resource assignment technique (CRAT) and a cooperation-driven population evolution method (CPEM). During evolution, CRAT evaluates each CTA’s validity and prioritizes resources for the more valid constraint-treating approaches (CTAs). To leverage each CTA’s strengths, CPEM co-evolves by recombining parents and diffusing offspring. Four distinct CTAs are selected to form a set for a competition and cooperation-driven evolutionary framework. To enhance search capability, an evolutionary strategy integrating historical and heuristic is incorporated into this framework. Numerical results on twenty-eight problems from IEEE CEC 2017 and an optimal AV parking problem show that CCDDE-IMCTAs achieve better solutions, robustness, faster convergence, smaller standard deviation, and less computation time than other state-of-the-art methods.

Details DOI

EAAI Journal 2024 Journal Article

Permanent magnet synchronous motor inter-turn short circuit diagnosis based on physical-data dual model under oil-drilling environment

  • Minglei Li
  • Yanfeng Geng
  • Weiliang Wang
  • Mengyu Tu
  • Xiang Wu

Inter-turn Short Circuit (ITSC) faults in Permanent Magnet Synchronous Motor (PMSM) have gained significant attention due to the growing demand for enhanced reliability and safety in actuation systems. This paper presents a adaptive fault diagnosis approach specifically designed for ITSC faults in PMSM used in oil-drilling applications, where sparse actual data and varying speed conditions pose considerable sparse training data and ITSC feature extraction challenges respectively. To address these challenges, we first construct a physical fault model of PMSM-ITSC and establish a simulation experiment platform to replicate downhole environments with high temperatures and rapidly changing PMSM speeds, ensuring a reliable data source for analysis. Subsequently, we propose a novel adaptive peak-to-peak self-finding method (APPS) that leverages frequency-domain prior knowledge to adaptively extract ITSC fault characteristics, even amidst drastic changes in PMSM speed. Furthermore, we introduce the time-sequence efficient moving window self-attention network (EMWSAN) data model for inferring the stator phase winding state, incorporating Half-sandwich and Cascaded windows group attention operations. This approach significantly reduces computational complexity and network parameters compared to traditional self-attention mechanisms. To expedite the fitting process, we apply transfer learning (TL) theory, transferring knowledge from the physical knowledge to the data model, enabling EMWSAN to be trained more efficiently. As a result, our proposed ITSC fault diagnosis scheme achieves an impressive 96. 72% classification accuracy, achieving existing state-of-the-art methods.

Details DOI

EAAI Journal 2022 Journal Article

Chance constrained dynamic optimization approach for single machine scheduling involving flexible maintenance, production, and uncertainty

  • Xiang Wu
  • Kanjian Zhang

Chance constraints are suitable for industrial process modeling under uncertain conditions, where constraints cannot be strictly satisfied or do not need to be fully satisfied. In this paper, a single machine scheduling problem involving flexible maintenance, production, and uncertainty is modeled as a chance constrained dynamic optimization problem (CCDOP). A novel method is proposed for transforming the CCDOP into an equivalent deterministic dynamic optimization problem (DOP) with fixed state jump times. Furthermore, by using the idea of l 1 penalty function and a smooth approximation technique, the resulting deterministic DOP becomes a smoothing penalty problem, which is a non-convex nonlinear parameter optimization problem (NNPOP) with simple bounds on the variables. To solve the NNPOP, a gradient-based stochastic search algorithm (GSSA) is developed based on a gradient-based adaptive search algorithm (GASA) and a novel stochastic search algorithm (NSSA). The convergence analysis result shows that the GSSA is a globally convergent algorithm. Finally, two numerical examples are used to illustrate the effectiveness of the proposed method. Numerical results show that the GSSA has excellent convergence behavior with robust computation feature, providing better results compared with the other typical methods.

Details DOI

NeurIPS Conference 2019 Conference Paper

Breaking the Glass Ceiling for Embedding-Based Classifiers for Large Output Spaces

  • Chuan Guo
  • Ali Mousavi
  • Xiang Wu
  • Daniel Holtmann-Rice
  • Satyen Kale
  • Sashank Reddi
  • Sanjiv Kumar

In extreme classification settings, embedding-based neural network models are currently not competitive with sparse linear and tree-based methods in terms of accuracy. Most prior works attribute this poor performance to the low-dimensional bottleneck in embedding-based methods. In this paper, we demonstrate that theoretically there is no limitation to using low-dimensional embedding-based methods, and provide experimental evidence that overfitting is the root cause of the poor performance of embedding-based methods. These findings motivate us to investigate novel data augmentation and regularization techniques to mitigate overfitting. To this end, we propose GLaS, a new regularizer for embedding-based neural network approaches. It is a natural generalization from the graph Laplacian and spread-out regularizers, and empirically it addresses the drawback of each regularizer alone when applied to the extreme classification setup. With the proposed techniques, we attain or improve upon the state-of-the-art on most widely tested public extreme classification datasets with hundreds of thousands of labels.

PDF Details

AAAI Conference 2019 Conference Paper

Disentangled Variational Representation for Heterogeneous Face Recognition

  • Xiang Wu
  • Huaibo Huang
  • Vishal M. Patel
  • Ran He
  • Zhenan Sun

Visible (VIS) to near infrared (NIR) face matching is a challenging problem due to the significant domain discrepancy between the domains and a lack of sufficient data for training cross-modal matching algorithms. Existing approaches attempt to tackle this problem by either synthesizing visible faces from NIR faces, extracting domain-invariant features from these modalities, or projecting heterogeneous data onto a common latent space for cross-modal matching. In this paper, we take a different approach in which we make use of the Disentangled Variational Representation (DVR) for crossmodal matching. First, we model a face representation with an intrinsic identity information and its within-person variations. By exploring the disentangled latent variable space, a variational lower bound is employed to optimize the approximate posterior for NIR and VIS representations. Second, aiming at obtaining more compact and discriminative disentangled latent space, we impose a minimization of the identity information for the same subject and a relaxed correlation alignment constraint between the NIR and VIS modality variations. An alternative optimization scheme is proposed for the disentangled variational representation part and the heterogeneous face recognition network part. The mutual promotion between these two parts effectively reduces the NIR and VIS domain discrepancy and alleviates over-fitting. Extensive experiments on three challenging NIR-VIS heterogeneous face recognition databases demonstrate that the proposed method achieves significant improvements over the state-of-the-art methods.

PDF Details

NeurIPS Conference 2019 Conference Paper

Dual Variational Generation for Low Shot Heterogeneous Face Recognition

  • Chaoyou Fu
  • Xiang Wu
  • Yibo Hu
  • Huaibo Huang
  • Ran He

Heterogeneous Face Recognition (HFR) is a challenging issue because of the large domain discrepancy and a lack of heterogeneous data. This paper considers HFR as a dual generation problem, and proposes a novel Dual Variational Generation (DVG) framework. It generates large-scale new paired heterogeneous images with the same identity from noise, for the sake of reducing the domain gap of HFR. Specifically, we first introduce a dual variational autoencoder to represent a joint distribution of paired heterogeneous images. Then, in order to ensure the identity consistency of the generated paired heterogeneous images, we impose a distribution alignment in the latent space and a pairwise identity preserving in the image space. Moreover, the HFR network reduces the domain discrepancy by constraining the pairwise feature distances between the generated paired heterogeneous images. Extensive experiments on four HFR databases show that our method can significantly improve state-of-the-art results. When using the generated paired images for training, our method gains more than 18\% True Positive Rate improvements over the baseline model when False Positive Rate is at $10^{-5}$.

PDF Details

IJCAI Conference 2019 Conference Paper

Neurons Merging Layer: Towards Progressive Redundancy Reduction for Deep Supervised Hashing

  • Chaoyou Fu
  • Liangchen Song
  • Xiang Wu
  • Guoli Wang
  • Ran He

Deep supervised hashing has become an active topic in information retrieval. It generates hashing bits by the output neurons of a deep hashing network. During binary discretization, there often exists much redundancy between hashing bits that degenerates retrieval performance in terms of both storage and accuracy. This paper proposes a simple yet effective Neurons Merging Layer (NMLayer) for deep supervised hashing. A graph is constructed to represent the redundancy relationship between hashing bits that is used to guide the learning of a hashing network. Specifically, it is dynamically learned by a novel mechanism defined in our active and frozen phases. According to the learned relationship, the NMLayer merges the redundant neurons together to balance the importance of each output neuron. Moreover, multiple NMLayers are progressively trained for a deep hashing network to learn a more compact hashing code from a long redundant code. Extensive experiments on four datasets demonstrate that our proposed method outperforms state-of-the-art hashing methods.

PDF Details

AAAI Conference 2018 Conference Paper

Adversarial Discriminative Heterogeneous Face Recognition

  • Lingxiao Song
  • Man Zhang
  • Xiang Wu
  • Ran He

The gap between sensing patterns of different face modalities remains a challenging problem in heterogeneous face recognition (HFR). This paper proposes an adversarial discriminative feature learning framework to close the sensing gap via adversarial learning on both raw-pixel space and compact feature space. This framework integrates cross-spectral face hallucination and discriminative feature learning into an endto-end adversarial network. In the pixel space, we make use of generative adversarial networks to perform cross-spectral face hallucination. An elaborate two-path model is introduced to alleviate the lack of paired images, which gives consideration to both global structures and local textures. In the feature space, an adversarial loss and a high-order variance discrepancy loss are employed to measure the global and local discrepancy between two heterogeneous distributions respectively. These two losses enhance domain-invariant feature learning and modality independent noise removing. Experimental results on three NIR-VIS databases show that our proposed approach outperforms state-of-the-art HFR methods, without requiring of complex network or large-scale training dataset.

PDF Details

AAAI Conference 2018 Conference Paper

Anti-Makeup: Learning A Bi-Level Adversarial Network for Makeup-Invariant Face Verification

  • Yi Li
  • Lingxiao Song
  • Xiang Wu
  • Ran He
  • Tieniu Tan

Makeup is widely used to improve facial attractiveness and is well accepted by the public. However, different makeup styles will result in significant facial appearance changes. It remains a challenging problem to match makeup and non-makeup face images. This paper proposes a learning from generation approach for makeup-invariant face verification by introducing a bi-level adversarial network (BLAN). To alleviate the negative effects from makeup, we first generate non-makeup images from makeup ones, and then use the synthesized nonmakeup images for further verification. Two adversarial networks in BLAN are integrated in an end-to-end deep network, with the one on pixel level for reconstructing appealing facial images and the other on feature level for preserving identity information. These two networks jointly reduce the sensing gap between makeup and non-makeup images. Moreover, we make the generator well constrained by incorporating multiple perceptual losses. Experimental results on three benchmark makeup face datasets demonstrate that our method achieves state-of-the-art verification accuracy across makeup status and can produce photo-realistic non-makeup face images.

PDF Details

AAAI Conference 2018 Conference Paper

Coupled Deep Learning for Heterogeneous Face Recognition

  • Xiang Wu
  • Lingxiao Song
  • Ran He
  • Tieniu Tan

Heterogeneous face matching is a challenge issue in face recognition due to large domain difference as well as insufficient pairwise images in different modalities during training. This paper proposes a coupled deep learning (CDL) approach for the heterogeneous face matching. CDL seeks a shared feature space in which the heterogeneous face matching problem can be approximately treated as a homogeneous face matching problem. The objective function of CDL mainly includes two parts. The first part contains a trace norm and a block-diagonal prior as relevance constraints, which not only make unpaired images from multiple modalities be clustered and correlated, but also regularize the parameters to alleviate overfitting. An approximate variational formulation is introduced to deal with the difficulties of optimizing low-rank constraint directly. The second part contains a cross modal ranking among triplet domain specific images to maximize the margin for different identities and increase data for a small amount of training samples. Besides, an alternating minimization method is employed to iteratively update the parameters of CDL. Experimental results show that CDL achieves better performance on the challenging CASIA NIR-VIS 2. 0 face recognition database, the IIIT-D Sketch database, the CUHK Face Sketch (CUFS), and the CUHK Face Sketch FERET (CUFSF), which significantly outperforms state-ofthe-art heterogeneous face recognition methods.

PDF Details

AAAI Conference 2017 Conference Paper

Learning Invariant Deep Representation for NIR-VIS Face Recognition

  • Ran He
  • Xiang Wu
  • Zhenan Sun
  • Tieniu Tan

Visual versus near infrared (VIS-NIR) face recognition is still a challenging heterogeneous task due to large appearance difference between VIS and NIR modalities. This paper presents a deep convolutional network approach that uses only one network to map both NIR and VIS images to a compact Euclidean space. The low-level layers of this network are trained only on large-scale VIS data. Each convolutional layer is implemented by the simplest case of maxout operator. The highlevel layer is divided into two orthogonal subspaces that contain modality-invariant identity information and modalityvariant spectrum information respectively. Our joint formulation leads to an alternating minimization approach for deep representation at the training time and an efficient computation for heterogeneous data at the testing time. Experimental evaluations show that our method achieves 94% verification rate at FAR=0. 1% on the challenging CASIA NIR-VIS 2. 0 face recognition dataset. Compared with state-of-the-art methods, it reduces the error rate by 58% only with a compact 64-D representation.

PDF Details

NeurIPS Conference 2017 Conference Paper

Multiscale Quantization for Fast Similarity Search

  • Xiang Wu
  • Ruiqi Guo
  • Ananda Theertha Suresh
  • Sanjiv Kumar
  • Daniel Holtmann-Rice
  • David Simcha
  • Felix Yu

We propose a multiscale quantization approach for fast similarity search on large, high-dimensional datasets. The key insight of the approach is that quantization methods, in particular product quantization, perform poorly when there is large variance in the norms of the data points. This is a common scenario for real- world datasets, especially when doing product quantization of residuals obtained from coarse vector quantization. To address this issue, we propose a multiscale formulation where we learn a separate scalar quantizer of the residual norm scales. All parameters are learned jointly in a stochastic gradient descent framework to minimize the overall quantization error. We provide theoretical motivation for the proposed technique and conduct comprehensive experiments on two large-scale public datasets, demonstrating substantial improvements in recall over existing state-of-the-art methods.

PDF Details

TIST Journal 2014 Journal Article

Object-Oriented Travel Package Recommendation

  • Chang Tan
  • Qi Liu
  • Enhong Chen
  • Hui Xiong
  • Xiang Wu

Providing better travel services for tourists is one of the important applications in urban computing. Though many recommender systems have been developed for enhancing the quality of travel service, most of them lack a systematic and open framework to dynamically incorporate multiple types of additional context information existing in the tourism domain, such as the travel area, season, and price of travel packages. To that end, in this article, we propose an open framework, the Objected-Oriented Recommender System (ORS), for the developers performing personalized travel package recommendations to tourists. This framework has the ability to import all the available additional context information to the travel package recommendation process in a cost-effective way. Specifically, the different types of additional information are extracted and uniformly represented as feature--value pairs. Then, we define the Object, which is the collection of the feature--value pairs. We propose two models that can be used in the ORS framework for extracting the implicit relationships among Objects. The Objected-Oriented Topic Model (OTM) can extract the topics conditioned on the intrinsic feature--value pairs of the Objects. The Objected-Oriented Bayesian Network (OBN) can effectively infer the cotravel probability of two tourists by calculating the co-occurrence time of feature--value pairs belonging to different kinds of Objects. Based on the relationships mined by OTM or OBN, the recommendation list is generated by the collaborative filtering method. Finally, we evaluate these two models and the ORS framework on real-world travel package data, and the experimental results show that the ORS framework is more flexible in terms of incorporating additional context information, and thus leads to better performances for travel package recommendations. Meanwhile, for feature selection in ORS, we define the feature information entropy, and the experimental results demonstrate that using features with lower entropies usually leads to better recommendation results.

Details DOI

YNIMG Journal 2007 Journal Article

Binding of verbal and spatial information in human working memory involves large-scale neural synchronization at theta frequency

  • Xiang Wu
  • Xiangchuan Chen
  • Zhihao Li
  • Shihui Han
  • Daren Zhang

Whether neural synchronization is engaged in binding of verbal and spatial information in working memory remains unclear. The present study analyzed oscillatory power and phase synchronization of electroencephalography (EEG) recorded from subjects performing a working memory task. Subjects were required to maintain both verbal (letters) and spatial (locations) information of visual stimuli while the verbal and spatial information were either bound or separate. We found that frontal theta power, and large-scale theta phase synchronization between bilateral frontal regions and between the left frontal and right temporal–parietal regions were greater for maintaining bound relative to separate information. However, the same effects were not observed in the gamma band. These results suggest that working memory binding involves large-scale neural synchronization at the theta band.

Details DOI
v2026.09.13