Search

Your search keyword '"Cai, Zixin"' showing total 131 results
Start Over Author "Cai, Zixin" Publication Year Range Last 10 years
131 results on '"Cai, Zixin"'

8. Hyperbolic Interaction Model For Hierarchical Multi-Label Classification

Author

Chen, Boli, Huang, Xin, Xiao, Lin, Cai, Zixin, and Jing, Liping

Subjects
Computer Science - Machine Learning, Computer Science - Computation and Language, Statistics - Machine Learning
Abstract

Different from the traditional classification tasks which assume mutual exclusion of labels, hierarchical multi-label classification (HMLC) aims to assign multiple labels to every instance with the labels organized under hierarchical relations. Besides the labels, since linguistic ontologies are intrinsic hierarchies, the conceptual relations between words can also form hierarchical structures. Thus it can be a challenge to learn mappings from word hierarchies to label hierarchies. We propose to model the word and label hierarchies by embedding them jointly in the hyperbolic space. The main reason is that the tree-likeness of the hyperbolic space matches the complexity of symbolic data with hierarchical structures. A new Hyperbolic Interaction Model (HyperIM) is designed to learn the label-aware document representations and make predictions for HMLC. Extensive experiments are conducted on three benchmark datasets. The results have demonstrated that the new model can realistically capture the complex data structures and further improve the performance for HMLC comparing with the state-of-the-art methods. To facilitate future research, our code is publicly available.

Published
2019

9. Optical synthetic sampling imaging: concept and an example of microscopy

Author

Peng, Junzheng, Yao, Manhong, Cai, Zixin, Qiu, Xue, Zhang, Zibang, Li, Shiping, and Zhong, Jingang

Subjects
Electrical Engineering and Systems Science - Image and Video Processing, Physics - Optics
Abstract

Digital two-dimensional (2D) spatial sampling devices (such as charge-coupled device) have been widely used in various imaging systems, especially in computational imaging systems. However, the undersampling of digital sampling devices is a problem that limits the resolution of the acquired images. In this study, we present a synthetic sampling imaging (SSI) concept to solve the undersampling problem. It combines the structured illumination system and conventional 2D image detection system to simultaneously sample the specimen from the illumination and the detection sides. Then, we synthesize the illumination sampling rate and the detection sampling rate to reconstruct a high sampling rate image. The concept of the proposed SSI is demonstrated by an example of microscopy. Experimental results confirm that the proposed method can double the sampling resolution of the microscope. The synthetic sampling scheme, where the sampling task is shared by the illumination and detection sides, provides insight for resolving the undersampling problem of the digital imaging system.

Published
2019
Full Text
View/download PDF

14. Spatio-Temporal Correlation Based Adaptive Feature Learning of Tracking Object

Author

GUO Mingzhe, CAI Zixin, WANG Xinyue, JING Liping, YU Jian

Subjects
spatio-temporal correlation, feature, tracking, object localization, object representation, Electronic computers. Computer science, QA75.5-76.95
Abstract

Object tracking has been a difficult problem in the field of vision in recent years. The core task is to continuously locate an object in video sequences and mark its location with bounding boxes. Most of the existing tracking methods use the idea of object detection, and separate the video sequence by frame to detect the target separately. Although this strategy makes full use of the current frame information, it ignores the spatio-temporal correlation information among frames. However, the spatio-temporal correlation information is the key of adapting to the change of the target??s appearance and fully representing the target. To solve this problem, this paper proposes a spatio-temporal siamese network (STSiam) based on spatio-temporal correlation. STSiam uses the spatio-temporal correlation information for target locating and real-time tracking in two stages: object localization and object repre-sentation. In the stage of object localization, STSiam adaptively captures the features of the target and its surroun-ding area, and updates the target matching template to ensure that it is not affected by appearance changes. In the stage of object representation, STSiam pays attention to the spatial correlation information between corresponding regions in different frames. By using the object localization, STSiam locates the target area and learns the target bounding box correction parameters to ensure that the bounding box fits the target as closely as possible. The model's network architecture is based on offline training, and it is no need to update model parameters during online tracking to ensure its real-time tracking speed. Extensive experiments on visual tracking benchmarks including OTB2015, VOT2016, VOT2018 and LaSOT demonstrate that STSiam achieves state-of-the-art performance in terms of accu-racy, robustness and speed compared with existing methods.

Published
2021
Full Text
View/download PDF

15. Meta-analysis of the association between sex hormones and pulmonary fibrosis

Author

Chen, Ying, Zhong, Jiaxin, Cai, Zixin, Xia, Zhenkun, Qing, Bei, Yuan, Yunchang, and Zhang, Jingjing

Abstract

ABSTRACTBackgroundThis study aimed to investigate the association between sex hormones and the risk of pulmonary fibrosis by conducting a meta-analysis of previously published studies.MethodsWe executed a comprehensive search of the PubMed, Embase, Cochrane Library, and Web of Science databases to locate pertinent studies published up to April 2024. We included studies that reported the association between sex hormones and the risk of pulmonary fibrosis. Standardized mean difference (SMD) with 95% confidence intervals (CIs) were calculated using a random-effects model.ResultsA total of 10 articles, encompassing 1371 patients, were finally incorporated in this meta-analysis. Based on the evaluation of the included studies, it was observed that the levels of dehydroepiandrosterone sulfate (DHEA-S) (pooled SMD: −0.72, 95% CI: −1.21 to −0.24, p < 0.001), testosterone (pooled SMD: −1.25, CI: −2.39 and −0.11, p < 0.001) and estrogen (pooled SMD: −0.56, 95% CI: −0.96 to −0.15, p < 0.001) were significantly lower in patients with pulmonary fibrosis, whereas the levels of luteinizing hormone (LH) remained unaffected. Publication bias was ruled out through funnel plots.ConclusionThis meta-analysis indicates that reduced levels of DHEA-S, testosterone, estrogen may serve as potential risk factors for pulmonary fibrosis. There is a pressing need for additional studies to confirm this association and explore the underlying biological mechanisms. Clinicians should recognize the potential influence of sex hormones in the etiology of pulmonary fibrosis and consider this aspect during the patient management process.

Published
2024
Full Text
View/download PDF

18. Imbalance Classification Based on Informative Instances Selection

Author

XU Jian, WANG Xinyue, CAI Zixin, SHEN Qihang, JING Liping

Subjects
class imbalance learning, classification, squeeze-and-excitation networks, cost-sensitive learning, Electronic computers. Computer science, QA75.5-76.95
Abstract

Class imbalance is a common challenge issue in practical classification problem for traditional models. Due to traditional learning algorithms can not sufficiently learn the hidden patterns from the minority classes and may be biased towards majority classes, thus minority instances are usually misclassified into majority instances. Moreover, redundant data and noise data in the dataset can also cause problems for the classifier. To deal with the above problems, this paper proposes a new imbalance classification framework SSIC. The framework fully considers the statistical properties of dataset, adaptively selects valuable instances from the different classes, and combines cost-sensitive learning to construct an imbalance classifier. Firstly, SSIC constructs several balanced data subsets by combining partial majority-class instances and all minority-class instances. On each subset, SSIC sufficiently takes advantage of the characteristics of data to extract the discriminative high-level features and adaptively selects the impor-tant samples, so that the redundant and noise data can be removed. Secondly, SSIC introduces a cost-sensitive support vector machine (SVM) by automatically assigning proper weight on each instance so that the minority class can be treated as equal as the majority class.

Published
2020
Full Text
View/download PDF

24. 基于超表面实现波长选择性波前整形

Author

Cai, Zixin, primary, He, Xin, additional, Liu, Xin, additional, Tu, Shijie, additional, Sun, Xinjie, additional, Beckett, Paul, additional, Dubey, Aditya, additional, Mitchell, Arnan, additional, Ren, Guanghui, additional, Liu, Xu, additional, and Hao, Xiang, additional

Published
2023
Full Text
View/download PDF

30. The adipocyte-enriched secretory protein tetranectin exacerbates type 2 diabetes by inhibiting insulin secretion from β cells

Author

Liu, Fen, primary, Cai, Zixin, additional, Yang, Yan, additional, Plasko, George, additional, Zhao, Piao, additional, Wu, Xiangyue, additional, Tang, Cheng, additional, Li, Dandan, additional, Li, Ting, additional, Hu, Shanbiao, additional, Song, Lei, additional, Yu, Shaojie, additional, Xu, Ran, additional, Luo, Hairong, additional, Fan, Libin, additional, Wang, Ersong, additional, Xiao, Zhen, additional, Ji, Yujiao, additional, Zeng, Rong, additional, Li, Rongxia, additional, Bai, Juli, additional, Zhou, Zhiguang, additional, Liu, Feng, additional, and Zhang, Jingjing, additional

Published
2022
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results