检索结果-内蒙古大学图书馆

Variational Gaussian Topic Model with Invertible Neural Projections

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Rui Zhou, Deyu Xiong, Yuxuan Huang, Haiping School of Computer Science Nanjing University of Posts and Telecommunications China Key Laboratory of Computer Network and Information Integration School of Computer Science and Engineering Ministry of Education Southeast University China Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks Nanjing China

Neural topic models have triggered a surge of interest in extracting topics from text automatically since they avoid the sophisticated derivations in conventional topic models. However, scarce neural topic models incorporate the word relatedness information captured in word embedding into the modeling process. To address this issue, we propose a novel topic modeling approach, called Variational Gaussian Topic Model (VaGTM). Based on the variational auto-encoder, the proposed VaGTM models each topic with a multivariate Gaussian in decoder to incorporate word relatedness. Furthermore, to address the limitation that pre-trained word embeddings of topicassociated words do not follow a multivariate Gaussian, Variational Gaussian Topic Model with Invertible neural Projections (VaGTM-IP) is extended from VaGTM. Three benchmark text corpora are used in experiments to verify the effectiveness of VaGTM and VaGTMIP. The experimental results show that VaGTM and VaGTM-IP outperform several competitive baselines and obtain more coherent topics. Copyright © 2021, The Authors. All rights reserved.

关键词： Gaussian distribution

A Dilated Transformer network for Time Series Anomaly Detection

学校读者我要写书评

暂无评论

A Dilated Transformer Network for Time Series Anomaly Detect...

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Bo Wu Zhenjie Yao Yanhui Tu Yixin Chen Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education Purple Mountain Laboratories Nanjing China Institute of Microelectronics Chinese Academy of Sciences Purple Mountain Laboratories Beijing China Shandong Future Network Research Institute Purple Mountain Laboratories Beijing China Washington University in St.Louis Purple Mountain Laboratories Washington USA

Unsupervised anomaly detection for time series has been an active research area due to its enormous potential for wireless network management. Existing works have made extraordinary progress in time series representation, reconstruction and forecasting. However, long-term temporal patterns prohibit the model from learning reliable dependencies. To this end, we propose a novel approach based on Transformer with dilated convolution for time anomaly detection. Specifically, we provide a dilated convolution module to extract long-term dependence features. Extensive experiments on various public benchmarks demonstrate that our method has achieved the state-of-the-art performance.

关键词： Convolution Wireless networks Time series analysis Benchmark testing Transformers Feature extraction Market research

Sequential-Scanning Dual-Energy CT Imaging Using High Temporal Resolution Image Reconstruction and Error-Compensated Material Basis Image Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Qiaoxin Chen, Ruifeng Wang, Peng Quan, Guotao Du, Yanfeng Liang, Dong Li, Yinsheng Research Center for Medical Artificial Intelligence Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Chinese Academy of Sciences Beijing China Laboratory of Image Science and Technology Key Laboratory of Computer Network and Information Integration Southeast University Nanjing China CT RPA Department United Imaging Healthcare Company Ltd. Shanghai China Research Center for Medical Artificial Intelligence Key Laboratory of Biomedical Imaging Science and System Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

Dual-energy computed tomography (DECT) has been widely used to obtain quantitative elemental composition of imaged subjects for personalized and precise medical diagnosis. Compared with DECT leveraging advanced X-ray source and/or detector technologies, the use of the sequential-scanning data acquisition scheme to implement DECT may make a broader impact on clinical practice because this scheme requires no specialized hardware designs and can be directly implemented into conventional CT systems. However, since the concentration of iodinated contrast agent in the imaged subject varies over time, sequentially scanned data sets acquired at two tube potentials are temporally inconsistent. As existing material basis image reconstruction approaches assume that the data sets acquired at two tube potentials are temporally consistent, the violation of this assumption results in inaccurate quantification of material concentration. In this work, we developed sequential-scanning DECT imaging using high temporal resolution image reconstruction and error-compensated material basis image generation, ACCELERATION in short, to address the technical challenge induced by temporal inconsistency of sequentially scanned data sets and improve quantification accuracy of material concentration in sequential-scanning DECT. ACCELERATION has been validated and evaluated using numerical simulation data sets generated from clinical human subject exams and experimental human subject studies. Results demonstrated the improvement of quantification accuracy and image quality using ACCELERATION. Copyright © 2024, The Authors. All rights reserved.

关键词： computerized tomography

Cone-Beam Computed Tomography Limited-Angle Image Reconstruction Using Planning CT-Guided Deep Diffusion Model

学校读者我要写书评

暂无评论

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Wu, Zhan Yang, Yang Hu, Dianlin Lyu, Tianling Li, Yinsheng Tang, Hao Shu, Huazhong Xi, Yan Zhang, Yi Chen, Yang Southeast University Laboratory of Image Science and Technology Nanjing210096 China Southeast University Ministry of Education Key Laboratory of Computer Network and Information Integration Nanjing210096 China Zhejiang University of Technology Academy for Advanced Interdisciplinary Science and Technology Hangzhou China Chinese Academy of Sciences Key Laboratory of Biomedical Imaging Science and System China Capital Medical University Beijing Jishuitan Hospital China First-Imaging Tech Shanghai China Sichuan University School of Cyber Science and Engineering Chengdu China Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Southeast University Ministry of Education Jiangsu Provincial Joint International Research Laboratory of Medical Information Processing Laboratory of Image Science and Technology the School of Computer Science and Engineering Nanjing210096 China

For existing image-guided radiation therapy systems (IGRT), the limited angular range of the cone-beam computed tomography (CBCT) scanning improves the compactness of the system and compatibility between CBCT and linear accelerator configurations. However, the limited-angle CBCT reconstruction under incomplete projection data causes severe artifacts in the reconstructed images, further having a negative impact on the registration accuracy in IGRT. In this paper, we propose a planning CT-guided deep diffusion model, termed LARGE-DiffNet, to improve imaging quality and registration accuracy for limited-angle CBCT image reconstruction. In stark contrast to existing approaches leveraging hand-crafted regularization, such as local smoothness and/or sparsity of the image, LARGE-DiffNet learns the distribution-level prior knowledge to describe the features of ideal CBCT images via a planning CT-guided diffusion model. The proposed LARGE-DiffNet is evaluated on the clinical CBCT dataset, and extensive experimental results demonstrate that the proposed method outperforms the state-of-the-art methods. The proposed LARGE-DiffNet technique has great potential in IGRT. © 1963-2012 IEEE.

关键词： Image reconstruction Computed tomography Planning Radiation therapy Diffusion models Biomedical imaging Image restoration Training Reconstruction algorithms Image quality

Reconstructing Quantitative Cerebral Perfusion Images Directly From Measured Sinogram Data Acquired Using C-arm Cone-Beam CT

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Haotian Chen, Ruifeng Yan, Jing Feng, Juan Xiang, Jun Chen, Yang Liang, Dong Li, Yinsheng Research Center for Medical Artificial Intelligence Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China Laboratory of Image Science and Technology Key Laboratory of Computer Network and Information Integration Southeast University Nanjing China X-Ray Department United Imaging Healthcare Limited Company Shanghai China Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications Southeast University Nanjing China Research Center for Medical Artificial Intelligence Key Laboratory of Biomedical Imaging Science and System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China

—To shorten the door-to-puncture time for better treating patients with acute ischemic stroke, it is highly desired to obtain quantitative cerebral perfusion images using C-arm conebeam computed tomography (CBCT) equipped in the interventional suite. However, limited by the slow gantry rotation speed, the temporal resolution and temporal sampling density of typical C-arm CBCT are much poorer than those of multi-detector-row CT in the diagnostic imaging suite. The current quantitative perfusion imaging includes two cascaded steps, time-resolved image reconstruction and perfusion parametric estimation. For time-resolved image reconstruction, technical challenge imposed by poor temporal resolution and poor sampling density causes inaccurate quantification of the temporal variation of cerebral artery and tissue attenuation values. For perfusion parametric estimation, it remains a technical challenge to appropriately design the handcrafted regularization for better solving the associated deconvolution problem. These two challenges together prevent from obtaining quantitatively accurate perfusion images using C-arm CBCT. The purpose of this work is to simultaneously address these two challenges via combining the two cascaded steps into a single joint optimization problem and reconstructing quantitative perfusion images directly from the measured sinogram data. In the developed direct cerebral perfusion parametric image reconstruction technique, TRAINER in short, the quantitative perfusion images have been represented as a subject-specific conditional generative model trained under the constraint of the time-resolved CT forward model, perfusion convolutional model, and the subject’s own measured sinogram data. Results shown in this paper demonstrated that using TRAINER, quantitative cerebral perfusion images can be accurately obtained using C-arm CBCT in the interventional suite. Copyright © 2024, The Authors. All rights reserved.

关键词： Medical imaging

Transfer synthetic over-sampling for class-imbalance learning with limited minority class data

学校读者我要写书评

暂无评论

Frontiers of computer Science 2019年第5期13卷 996-1009页

作者： Xu-Ying LIU Sheng-Tao WANG Min-Ling ZHANG School of Computer Science and Engineering Southeast University Nanjing 210096 China Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education Nanjing 210096 China Collaborative Innovation Center for Wireless Communications Technology Nanjing 210096 China

The problem of limited minority class data is encountered in manyclass imbalanced applications, but has received little attention. Synthetic over-sampling, as popular class-imbalance learning methods, could introduce much noise when minority class has limited data since the synthetic samples are not i.i.d. samples of minority class. Mos t sophisticated synthetic sampling methods tackle this problem by denoising or generating samples more consistent with ground-truth data distribution. But their assumptions about true noise or ground-truth data distribution may not hold. To adapt synthetic sampling to the problem of limited minority class data, the proposed Traso framework treats synthetic minority class samples as an additional data source, and exploits transfer learning to transfer knowledge from them to minority class. As an implementation, TrasoBoost method firstly generates synthetic samples to balance class sizes. Then in each boosting iteration, the weights of synthetic samples and original data decrease and increase respectively when being misclassified, and remain unchanged otherwise. The misclassified synthetic samples are potential noise, and thus have smaller influence in the following iterations. Besides, the weights of minority class instances have greater change than those of majority class instances to be more influential. And only original data are used to estimate error rate to be immune from noise. Finally, since the synthetic samples are highly related to minority class, all of the weak learners are aggregated for prediction. Experimental results show TrasoBoost outperforms many popular class-imbalance learning methods.

关键词： machine learning data mining class imbalance over sampling boosting transfer learning

A study on improved mrmf algorithm on uncertain graph 2nd

学校读者我要写书评

暂无评论

A study on improved mrmf algorithm on uncertain graph

2nd International Cognitive Cities Conference, IC3 2019

作者： Wen, Kejie Yang, Juan Liu, Wei Yang, Yanchun Zhang, Baili School of Computer Science and Engineering Southeast University Nanjing211189 China Key Laboratory of Computer Network and Information Integration of Ministry of Education Southeast University Nanjing211189 China Research Center for Judicial Big Data Supreme Count of China Nanjing211189 China

ISBN: (纸本)9789811561122

With the increased size and density of graph, the performance of the space decomposition-based algorithm (SDBA) decreases rapidly in terms of the most reliable maximum flow problem (MRMF). To solve this problem, the double filter space decomposition-based algorithm (DF-SDBA) is proposed by applying the probability filter and the cut-set filter. First, among the intervals to be processed in the process of space decomposition, DF-SDBA algorithm uses the probability constraint to filter out the intervals which own smaller probability of the lower boundary distribution than the current maximum flow, reducing the number of iterations effectively. Then the cut-set constraint is applied to unspecified intervals. According to the rule that all the edges in the cut-sets must exist in the upper bound of each interval, the unspecified intervals are filtered out, cutting down the number of partition intervals once again. The experimental results reveal that compared with the SDBA algorithm, the DF-SDBA algorithm can reduce the number of the intervals needed to be divided can obtain better performance. © Springer Nature Singapore Pte Ltd 2020.

关键词： Flow graphs

Neural Topic Modeling by Incorporating Document Relationship Graph

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhou, Deyu Hu, Xuemeng Wang, Rui School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University China

Graph Neural networks (GNNs) that capture the relationships between graph nodes via message passing have been a hot research direction in the natural language processing community. In this paper, we propose Graph Topic Model (GTM), a GNN based neural topic model that represents a corpus as a document relationship graph. Documents and words in the corpus become nodes in the graph and are connected based on document-word co-occurrences. By introducing the graph structure, the relationships between documents are established through their shared words and thus the topical representation of a document is enriched by aggregating information from its neighboring nodes using graph convolution. Extensive experiments on three datasets were conducted and the results demonstrate the effectiveness of the proposed approach. Copyright © 2020, The Authors. All rights reserved.

关键词： Message passing

Neural topic modeling with cycle-consistent adversarial training

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Hu, Xuemeng Wang, Rui Zhou, Deyu Xiong, Yuxuan School of Computer Science and Engineering Key Laboratory of Computer Network and Information Integration Ministry of Education Southeast University China

Advances on deep generative models have attracted significant research interest in neural topic modeling. The recently proposed Adversarial-neural Topic Model models topics with an adversarially trained generator network and employs Dirichlet prior to capture the semantic patterns in latent topics. It is effective in discovering coherent topics but unable to infer topic distributions for given documents or utilize available document labels. To overcome such limitations, we propose Topic Modeling with Cycle-consistent Adversarial Training (ToMCAT) and its supervised version sToMCAT. ToMCAT employs a generator network to interpret topics and an encoder network to infer document topics. Adversarial training and cycle-consistent constraints are used to encourage the generator and the encoder to produce realistic samples that coordinate with each other. sToMCAT extends ToMCAT by incorporating document labels into the topic modeling process to help discover more coherent topics. The effectiveness of the proposed models is evaluated on unsupervised/supervised topic modeling and text classification. The experimental results show that our models can produce both coherent and informative topics, outperforming a number of competitive baselines. Copyright © 2020, The Authors. All rights reserved.

关键词： Semantics