检索结果-内蒙古大学图书馆

Multi-view Contrastive Learning for Medical Question Summarization

学校读者我要写书评

暂无评论

Multi-view Contrastive Learning for Medical Question Summari...

International Conference on computer Supported Cooperative Work in Design

作者： Sibo Wei Xueping Peng Hongjiao Guan Lina Geng Ping Jian Hao Wu Wenpeng Lu Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Key Laboratory of Computing Power Network and Information Security Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Australian Artificial Intelligence Institute University of Technology Sydney Australia Department of Blood Purification Qilu Hospital of Shandong University Jinan China School of Computer Science and Technology Beijing Institute of Technology Beijing China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Most Seq2Seq neural model-based medical question summarization (MQS) systems have a severe mismatch between training and inference, i.e., exposure bias. However, this problem remains unexplored in the MQS task. To bridge this research gap and alleviate the problem of exposure bias, we propose a novel re-ranking training framework for MQS called Multi-view Contrastive Learning (MvCL). MvCL simultaneously considers the similarity scores between medical questions and candidate summaries as well as the average similarity scores between candidate summaries and other candidates within the same group, and utilizes contrastive learning to optimize the model’s ranking ability. Additionally, we propose a new multilevel inference approach to adapt to this training strategy. The approach first filters out candidate summaries that are dissimilar to the original medical question, and then selects the summary with the highest average similarity to other candidate summaries from the remaining candidates as the final output. We conducted extensive experiments, and the results demonstrate that our proposed MvCL framework achieves state-of-the-art results on the majority of evaluation metrics across four datasets. 1

关键词： Training Measurement Bridges Adaptation models Limiting Filters Federated learning

Learnable Graph Convolutional Network and Feature Fusion for Multi-view Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Zhaoliang Fu, Lele Yao, Jie Guo, Wenzhong Plant, Claudia Wang, Shiping College of Computer and Data Science Fuzhou University Fuzhou350116 China Fujian Provincial Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou350116 China Faculty of Computer Science University of Vienna Vienna1090 Austria Research Network Data Science @ Uni Vienna University of Vienna Vienna1090 Austria

In practical applications, multi-view data depicting objectives from assorted perspectives can facilitate the accuracy increase of learning algorithms. However, given multi-view data, there is limited work for learning discriminative node relationships and graph information simultaneously via graph convolutional network that has drawn the attention from considerable researchers in recent years. Most of existing methods only consider the weighted sum of adjacency matrices, yet a joint neural network of both feature and graph fusion is still under-explored. To cope with these issues, this paper proposes a joint deep learning framework called Learnable Graph Convolutional Network and Feature Fusion (LGCN-FF), consisting of two stages: feature fusion network and learnable graph convolutional network. The former aims to learn an underlying feature representation from heterogeneous views, while the latter explores a more discriminative graph fusion via learnable weights and a parametric activation function dubbed Differentiable Shrinkage Activation (DSA) function. The proposed LGCN-FF is validated to be superior to various state-of-the-art methods in multi-view semi-supervised classification. Copyright © 2022, The Authors. All rights reserved.

关键词： Learning algorithms

Network Parameter Generation for One-Shot Object Detection

学校读者我要写书评

暂无评论

Network Parameter Generation for One-Shot Object Detection

IEEE International Conference on Unmanned Systems (ICUS)

作者： Kang Wang Jiajia Zhao Yang Wang Shuigeng Zhou Shanghai Key Lab of Intelligent Information Processing and School of Computer Science Fudan University Shanghai China Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory Beijing Electro-Mechanical Engineer Institute Beijing China School of Information Engineering Nanchang University China

ISBN: (纸本)9781728137933

Recent object detection models have achieved satisfactory performance by deep learning with large-scale annotated datasets. However, these models often perform poorly when the training examples are not sufficient enough, especially for one-shot object detection where there is only one training example of each target class. In this paper, we propose a network parameter generation framework for effective one-shot object detection. By leveraging the semantic relationship between the source-domain and the target-domain (new classes), our framework can generate network parameters for the new classes directly. Furthermore, we develop two implementation schemes for this framework: weighting-based parameter generation and regression-based parameter generation. These two schemes use different information to construct the relationship and transfer knowledge from the source-domain to the target-domain. Experiments on the PASCAL-OSCD benchmark show that our methods can significantly boost the performance of one-shot object detection.

关键词： Object detection Detectors Head Training Feature extraction Task analysis Training data

A Novel Text Matching Model Based on Multilayer Coding and Feature Enhancement

学校读者我要写书评

暂无评论

A Novel Text Matching Model Based on Multilayer Coding and F...

International Conference on computer Supported Cooperative Work in Design

作者： Yu Zhang Jianqiang Zhang Gongpeng Song Qin Lu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Shandong Branch of China Mobile Communication Group Design Institute Co Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

With the rapid development of Natural Language processing (NLP), text matching has become the basis of many downstream tasks in NLP, and the study of text matching is of great research significance for solving tasks such as question and answer and information retrieval in NLP. Most of the current text matching methods tend to have problems such as mismatch of grammatical structures and insufficient interaction information in sentences. In order to solve the problems of insufficient interaction information and lack of features and ability to capture keyword and sequence information in text matching, this paper proposes a text matching method based on multi-layer coding and soft attention mechanism. The method first embeds the text and sends it to a gating module for processing, then sends the processed result to a module containing a combination of multilayer coding and soft attention mechanism for further operations such as multiple alignment, and finally sends it to a classifier containing a three-layer fully-connected network for predicting whether the input text pairs match or not. The two modules proposed in this paper are practically feasible, and comparison and ablation experiments have been conducted on the publicly available datasets LCQMC dataset and BQ dataset, and the experimental results show that the two modules improve the accuracy of text matching by 2.44% and 1.02%, respectively.

关键词： Attention mechanisms Accuracy Computational modeling Impedance matching Nonhomogeneous media information retrieval Encoding

The aging effect in evolving scientific citation networks

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Hu, Feng Ma, Lin Zhan, Xiu-Xiu Zhou, Yinzuo Liu, Chuang Zhao, Haixing Zhang, Zi-Ke School of Computer Qinghai Normal University Xining810016 China Key Laboratory of Tibetan Information Processing Ministry of Education Xining810008 China Tibetan Information Processing and Machine Translation Key Laboratory of Qinghai Province Xining810008 China Alibaba Research Center for Complexity Sciences Hangzhou Normal University Hangzhou311121 China Faculty of Electrical Engineering Mathematics and Computer Science Delft University of Technology Delft2628 CD Netherlands College of Media and International Culture Zhejiang University Hangzhou310058 China

The study of citation networks is of interest to the scientific community. However, the underlying mechanism driving individual citation behavior remains imperfectly understood, despite the recent proliferation of quantitative research methods. Traditional network models normally use graph theory to consider articles as nodes and citations as pairwise relationships between them. In this paper, we propose an alternative evolutionary model based on hypergraph theory in which one hyperedge can have an arbitrary number of nodes, combined with an aging effect to reflect the temporal dynamics of scientific citation behavior. Both theoretical approximate solution and simulation analysis of the model are developed and validated using two benchmark datasets from different disciplines, i.e. publications of the American Physical Society (APS) and the Digital Bibliography & Library Project (DBLP). Further analysis indicates that the attraction of early publications will decay exponentially. Moreover, the experimental results show that the aging effect indeed has a significant influence on the description of collective citation patterns. Shedding light on the complex dynamics driving these mechanisms facilitates the understanding of the laws governing scientific evolution and the quantitative evaluation of scientific outputs. © 2021, CC BY.

关键词： Graph theory

Research on Assistant Diagnostic Method of TCM Based on BERT

学校读者我要写书评

暂无评论

Research on Assistant Diagnostic Method of TCM Based on BERT

IEEE International Symposium on information (IT) in Medicine and Education, ITME

作者： Chuanjie Xu Feng Yuan Shouqiang Chen Shandong Provincial Key Laboratory for Novel Distributed Computer Software Technology Jinan China School of Information Engineering Shandong Management University Jinan China Center of Hear of the Second Affiliated Hospital of Shandong University of Traditional Chinese Medicine Jinan China

ISBN: (纸本)9781728194202

Traditional Chinese medicine (TCM) auxiliary diagnosis is a systematic diagnosis platform that uses computer modeling technology to assist TCM doctors in recording diseases, providing on time diagnoses, writing prescriptions, performing tele-medicine, and supporting medical teaching. This study proposes a Bidirectional Encoder Representations from Transformers TCM auxiliary diagnosis model using 20,000 items of TCM records. These records were collected from the outpatient clinic of the Second Affiliated Hospital of Shandong University of Traditional Chinese Medicine. Specifically, our model aims to predict final diagnosis while taking TCM symptoms as inputs; for example, when we input relief of chest tightness but persistent tiredness and sluggishness, the model provides a diagnosis of chest paralysis. An experiment was conducted on these real-world Chinese medical data. Results show that our model achieves state-of-the-art performance. Hence, our proposed model can effectively use the information from the four diagnostic procedures in the TCM text.

关键词： Systematics Computational modeling Education Predictive models Writing Transformers Data models

Cross-directional Consistency Network with Adaptive Layer Normalization for Multi-spectral Vehicle Re-identification and A High-quality Benchmark

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zheng, Aihua Zhu, Xianpeng Ma, Zhiqi Li, Chenglong Tang, Jin Ma, Jixin Information Materials and Intelligent Sensing Laboratory of Anhui Province Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Artificial Intelligence Anhui University Hefei230601 China Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China School of Computing and Mathematical Sciences University of Greenwich LondonSE10 9LS United Kingdom

To tackle the challenge of vehicle re-identification (Re-ID) in complex lighting environments and diverse scenes, multi-spectral sources like visible and infrared information are taken into consideration due to their excellent complementary advantages. However, multi-spectral vehicle Re-ID suffers cross-modality discrepancy caused by heterogeneous properties of different modalities as well as a big challenge of the diverse appearance with different views in each identity. Meanwhile, diverse environmental interference leads to heavy sample distributional discrepancy in each modality. In this work, we propose a novel cross-directional consistency network (CCNet) to simultaneously overcome the discrepancies from both modality and sample aspects. In particular, we design a new cross-directional center loss (Lcdc) to pull the modality centers of each identity close to mitigate cross-modality discrepancy, while the sample centers of each identity close to alleviate the sample discrepancy. Such a strategy can generate discriminative multi-spectral feature representations for vehicle Re-ID. In addition, we design an adaptive layer normalization unit (ALNU) to dynamically adjust individual feature distribution to handle distributional discrepancy of intra-modality features for robust learning. To provide a comprehensive evaluation platform, we create a high-quality RGB-NIR-TIR multi-spectral vehicle Re-ID benchmark (MSVR310), including 310 different vehicles from a broad range of viewpoints, time spans and environmental complexities. Comprehensive experiments on both created and public datasets demonstrate the effectiveness of the proposed approach comparing to the state-of-the-art methods. The dataset and code will be released for free academic usage at https://***/superlollipop123/Cross-directional-Center-Network-and-MSVR310. © 2022, CC BY-NC-SA.

关键词： Vehicles

Five A+Network: You Only Need 9K Parameters for Underwater Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jiang, Jingxia Ye, Tian Bai, Jinbin Chen, Sixiang Chai, Wenhao Jun, Shi Liu, Yun Chen, Erkang School of Ocean Information Engineering Jimei University Xiamen China Department of Computer Science National University of Singapore Singapore ZJU-UIUC Institute Zhejiang University Hanzhou China The School of Information Science and Engineering Xinjiang University Uruqmi China College of Artificial Intelligence Southwest University Chongqing China Fujian Provincial Key Laboratory of Oceanic Information Perception and Intelligent Processing China

A lightweight underwater image enhancement network is of great significance for resource-constrained platforms, but balancing model size, computational efficiency, and enhancement performance has proven difficult for previous approaches. In this work, we propose the Five A+Network (FA+Net), a highly efficient and lightweight real-time underwater image enhancement network with only ∼ 9k parameters and ∼ 0.01s processing time. The FA+Net employs a two-stage enhancement structure. The strong prior stage aims to decompose challenging underwater degradations into sub-problems, while the fine-grained stage incorporates multi-branch color enhancement module and pixel attention module to amplify the network's perception of details. To the best of our knowledge, FA+Net is the only network with the capability of real-time enhancement of 1080P images. Thorough extensive experiments and comprehensive visual comparison, we show that FA+Net outperforms previous approaches by obtaining state-of-the-art performance on multiple datasets while significantly reducing both parameter count and computational complexity. The code is open source at https://***/Owen718/FiveAPlusNetwork. Copyright © 2023, The Authors. All rights reserved.

关键词： Computational efficiency

EQ-Net: Elastic Quantization Neural Networks

学校读者我要写书评

暂无评论

EQ-Net: Elastic Quantization Neural Networks

International Conference on computer Vision (ICCV)

作者： Ke Xu Lei Han Ye Tian Shangshang Yang Xingyi Zhang School of Artificial Intelligence Anhui University Hefei China Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Hefei China School of Computer Science and Technology Anhui University Hefei China Institutes of Physical Science and Information Technology Anhui University Hefei China

Current model quantization methods have shown their promising capability in reducing storage space and computation complexity. However, due to the diversity of quantization forms supported by different hardware, one limitation of existing solutions is that usually require repeated optimization for different scenarios. How to construct a model with flexible quantization forms has been less studied. In this paper, we explore a one-shot network quantization regime, named Elastic Quantization Neural Networks (EQ-Net), which aims to train a robust weight-sharing quantization supernet. First of all, we propose an elastic quantization space (including elastic bit-width, granularity, and symmetry) to adapt to various mainstream quantitative forms. Secondly, we propose the Weight Distribution Regularization Loss (WDR-Loss) and Group Progressive Guidance Loss (GPG-Loss) to bridge the inconsistency of the distribution for weights and output logits in the elastic quantization space gap. Lastly, we incorporate genetic algorithms and the proposed Conditional Quantization-Aware Accuracy Predictor (CQAP) as an estimator to quickly search mixed-precision quantized neural networks in supernet. Extensive experiments demonstrate that our EQ-Net is close to or even better than its static counterparts as well as state-of-the-art robust bit-width methods. Code can be available at https://***/xuke225/EQ-Net.

关键词：