检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Bao, Shilong Xu, Qianqian Yang, Zhiyong He, Yuan Cao, Xiaochun Huang, Qingming State Key Laboratory of Information Security Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China School of Computer Science and Tech. University of Chinese Academy of Sciences China Alibaba Group China School of Cyber Science and Technology Shenzhen Campus Sun Yat-sen University China Key Laboratory of Big Data Mining and Knowledge Management CAS China Peng Cheng Laboratory China

Collaborative Metric Learning (CML) has recently emerged as a popular method in recommendation systems (RS), closing the gap between metric learning and Collaborative Filtering. Following the convention of RS, existing methods exploit unique user representation in their model design. This paper focuses on a challenging scenario where a user has multiple categories of interests. Under this setting, we argue that the unique user representation might induce preference bias, especially when the item category distribution is imbalanced. To address this issue, we propose a novel method called Diversity-Promoting Collaborative Metric Learning (DPCML), with the hope of considering the commonly ignored minority interest of the user. The key idea behind DPCML is to include a multiple set of representations for each user in the system. Based on this embedding paradigm, user preference toward an item is aggregated from different embeddings by taking the minimum item-user distance among the user embedding set. Furthermore, we observe that the diversity of the embeddings for the same user also plays an essential role in the model. To this end, we propose a diversity control regularization term to accommodate the multi-vector representation strategy better. Theoretically, we show that DPCML could generalize well to unseen test data by tackling the challenge of the annoying operation that comes from the minimum value. Experiments over a range of benchmark datasets speak to the efficacy of DPCML. © 2022, CC BY-NC-SA.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Msa-Net: A Medical Image Segmentation Network Based on Spatial Pyramid and Attention Mechanism

SSRN

引用

SSRN 2023年

作者： Feng, Yuncong Zhu, Xiaoyan Li, Yang Zhang, Xiaoli Lu, Huimin College of Computer Science and Engineering Changchun University of Technology Jilin Changchun130012 China Artificial Intelligence Research Institute Changchun University of Technology Jilin Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Jilin Changchun130012 China Academy for Advanced Interdisciplinary Studies Northeast Normal University Jilin Changchun130024 China Jilin Provincial Smart Health Joint Innovation Laboratory for the New Generation of AI 3000 North Yuanda Avenue Gaoxin North District Changchun130102 China

Medical image segmentation is an important way to assist doctors to accurately diagnose diseases. However, semantic features are difficult to be fully extracted due to the complexity of medical image lesion tissue features (e.g., size, location, and morphology) and the high similarity between the background and the target region. To tackle these problems, we propose a novel medical image segmentation network MSA-Net based on the spatial pyramid and attention mechanism. This technique facilitates the extraction of multi-scale spatial information and the establishment of channel dependencies within images. By incorporating efficient pyramid attention and channel-spatial dual attention modules, the network autonomously focuses on lesion regions while suppressing background interference. Additionally, the SPCA (Spatial Pyramid-Coordinate Attention) module is integrated at the bottleneck layer, enabling the assimilation of larger receptive field information and prioritizing relevant features. This augmentation improves the network's ability to gather both location and channel-based insights. We proceed to evaluate our model qualitatively and quantitatively across four public datasets: ISIC2018, Lung segmentation, Kvasir-SEG, and ISLES2022. The results substantiate the superior performance of our proposed model. © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Distilling Autoregressive Models to Obtain High-Performance Non-Autoregressive Solvers for Vehicle Routing Problems with Faster Inference Speed

arXiv

引用

arXiv 2023年

作者： Xiao, Yubin Wang, Di Li, Boyang Wang, Mingzhao Wu, Xuan Zhou, Changliang Zhou, You Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education College of Computer Science and Technology Jilin University China Joint NTU-UBC Research Centre of Excellence in Active Living for the Elderly Nanyang Technological University Singapore WeBank-NTU Joint Research Institute on Fintech Nanyang Technological University Singapore School of Computer Science and Engineering Nanyang Technological University Singapore School of System Design and Intelligent Manufacturing Southern University of Science and Technology China

Neural construction models have shown promising performance for Vehicle Routing Problems (VRPs) by adopting either the Autoregressive (AR) or Non-Autoregressive (NAR) learning approach. While AR models produce high-quality solutions, they generally have a high inference latency due to their sequential generation nature. Conversely, NAR models generate solutions in parallel with a low inference latency but generally exhibit inferior performance. In this paper, we propose a generic Guided Non-Autoregressive knowledge Distillation (GNARKD) method to obtain high-performance NAR models having a low inference latency. GNARKD removes the constraint of sequential generation in AR models while preserving the learned pivotal components in the network architecture to obtain the corresponding NAR models through knowledge distillation. We evaluate GNARKD by applying it to three widely adopted AR models to obtain NAR VRP solvers for both synthesized and real-world instances. The experimental results demonstrate that GNARKD significantly reduces the inference time (4∼5× faster) with acceptable performance drop (2∼3%). To the best of our knowledge, this study is first-of-its-kind to obtain NAR VRP solvers from AR ones through knowledge distillation. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Novel trusted authentication and evaluation framework in VANET

引用

Tongxin Xuebao/Journal on Communications 2009年第1 A期30卷 107-113页

作者： Wu, Jing Liu, Yan-Heng Wang, Jian Li, Wei-Ping College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China Changchun Jilin University Information Technologies Co. Ltd. Changchun 130012 China

Vehicular ad hoc networks (VANET) have many new characteristics unlike tranditional network, such as centriclessness, mobility and multi-hop transmission, which invalidate conventional key managements. The fast development and successful application of trusted computing technology provide a potential solution to achieve effective authentication and trust evaluation. According to the VANET with self-organization, a new authentication and trust evaluation framework was proposed based on trusted computing technology, the concepts of "trusted attribute" and "trusted grade" were introduced, and the key management architecture as well as authentication process of mobile node were explained in detail. Moreover, the potential real application of the presented framework was discussed, and the trust evaluation method and its argumentation in the presented framework were given.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Fully-convolutional intensive feature flow neural network for text recognition

arXiv

引用

arXiv 2019年

作者： Zhang, Zhao Tang, Zemin Zhang, Zheng Wang, Yang Qin, Jie Wang, Meng School of Computer Science and Technology Soochow University China Key Laboratory of Knowledge Engineering with Big Data Ministry of Education School of Computer and Information Hefei University of Technology Hefei China Bio-Computing Research Center Harbin Institute of Technology Shenzhen518055 China Inception Institute of Artificial Intelligence Abu Dhabi United Arab Emirates

The Deep Convolutional Neural Networks (CNNs) have obtained a great success for pattern recognition, such as recognizing the texts in images. But existing CNNs based frameworks still have several drawbacks: 1) the traditaional pooling operation may lose important feature information and is unlearnable;2) the traditional convolution operation optimizes slowly and the hierarchical features from different layers are not fully utilized. In this work, we address these problems by developing a novel deep network model called Fully-Convolutional Intensive Feature Flow Neural Network (IntensiveNet). Specifically, we design a further dense block called intensive block to extract the feature information, where the original inputs and two dense blocks are connected tightly. To encode data appropriately, we present the concepts of dense fusion block and further dense fusion operations for our new intensive block. By adding short connections to different layers, the feature flow and coupling between layers are enhanced. We also replace the traditional convolution by depthwise separable convolution to make the operation efficient. To prevent important feature information being lost to a certain extent, we use a convolution operation with stride 2 to replace the original pooling operation in the customary transition layers. The recognition results on large-scale Chinese string and MNIST datasets show that our IntensiveNet can deliver enhanced recognition results, compared with other related deep models. Copyright © 2019, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

CLDG: Contrastive Learning on Dynamic Graphs

arXiv

引用

arXiv 2024年

作者： Xu, Yiming Shi, Bin Ma, Teng Dong, Bo Zhou, Haoyi Zheng, Qinghua Department of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University China Department of Distance Education Xi’an Jiaotong University China School of Software Beihang University China Advanced Innovation Center for Big Data and Brain Computing Beihang University China

The graph with complex annotations is the most potent data type, whose constantly evolving motivates further exploration of the unsupervised dynamic graph representation. One of the representative paradigms is graph contrastive learning. It constructs self-supervised signals by maximizing the mutual information between the statistic graph’s augmentation views. However, the semantics and labels may change within the augmentation process, causing a significant performance drop in downstream tasks. This drawback becomes greatly magnified on dynamic graphs. To address this problem, we designed a simple yet effective framework named CLDG. Firstly, we elaborate that dynamic graphs have temporal translation invariance at different levels. Then, we proposed a sampling layer to extract the temporally-persistent signals. It will encourage the node to maintain consistent local and global representations, i.e., temporal translation invariance under the timespan views. The extensive experiments demonstrate the effectiveness and efficiency of the method on seven datasets by outperforming eight unsupervised state-of-the-art baselines and showing competitiveness against four semi-supervised methods. Compared with the existing dynamic graph method, the number of model parameters and training time is reduced by an average of 2,001.86 times and 130.31 times on seven datasets, respectively. The code and data are available at: https://***/yimingxu24/CLDG. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

CLDG: Contrastive Learning on Dynamic Graphs

CLDG: Contrastive Learning on Dynamic Graphs

引用

International Conference on Data engineering

作者： Yiming Xu Bin Shi Teng Ma Bo Dong Haoyi Zhou Qinghua Zheng Department of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University China Department of Distance Education Xi’an Jiaotong University China School of Software Beihang University China Advanced Innovation Center for Big Data and Brain Computing Beihang University China

关键词：

来源：评论

学校读者我要写书评

暂无评论

Human Eyes Move to the Target Earlier When Performing an Aiming Task with Increasing Difficulties

SSRN

引用

SSRN 2022年

作者： Liu, Xin Zhang, Yao Jiang, Xianta Zheng, Bin School of Computer and Communication Engineering University of Science and Technology Beijing Beijing100083 China Surgical Simulation Research Laboratory Department of Surgery University of Alberta EdmontonABT6G 2E1 Canada Department of Computing Science Memorial University of Newfoundland NewfoundlandA1B 3X5 Canada Beijing Key Laboratory of Knowledge Engineering for Materials Science Beijing100083 China

We examined whether human operators move their eyes earlier to a target before hands when the level of task difficulty increases. We hypothesized that subjects would perform less proactive eye movements in the difficult task than in the easy one, as they would need to focus on their current hand movements. Sixteen university students were recruited to perform the aiming and touching task reciprocally on three paired targets (circles) differing in sizes and distances, while had their eye movements tracked. The movement time, the early eye engagement time (EEET), and the number of eye adjustments were recorded. The EEET was defined as the time that a subject’s eyes fix on a target to the moment that the tool reaches out for it. The movement time increased as the index of difficulty (ID) value increased, echoing with the Fitts’ Law prediction. When aiming to a target with higher ID, subjects’ EEET was longer comparing to when reaching for a target with lower ID. Participants reduced movement speed to give themselves a longer time in searching visually for the target information before moving their hands. Results suggested that a proportional relationship may exist between task difficulty and the early eye engagement time. Subjects also performed an increasing number of eye adjustments over the course of moving from the easy to the hard target. Future research is needed to examine eye hand coordination under the regulation of Fitts’ Law. © 2022, The Authors. All rights reserved.

关键词： Eye tracking

来源：评论

学校读者我要写书评

暂无评论

Group-attention Based Neural Machine Translation

引用

IOP Conference Series: Materials Science and engineering 2020年第2期782卷

作者： Renyu Hu Hao Xu Yang Xiao Chenjun Wu Haiyang Jia College of Computer Science and Technology Jilin University Changchun 130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China

Machine translation is a classic problem in natural language process (NLP). Recent years, the encoder and decoder through an attention mechanism has become a trend. Google proposed a new simple network architecture, the Transformer using attention mechanisms only rather than CNN or RNN in 2017. However, it may lose some important information (e.g., grammatical component, etc) when using attention mechanism for whole *** propose a new brand model based on transformer using Group attention layers and group position embedding. It absorbs the features of Group-CNN combines the algorithm in computer vision (CV) and NLP. The model not only pays more attention to the ingredients (e.g., subject, predicate and adverbial, etc), but also enhances the connection of phrase. It outperforms SofA Transformer in using more syntactic information.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Learning Structured Twin-Incoherent Twin-Projective Latent Dictionary Pairs for Classification

Learning Structured Twin-Incoherent Twin-Projective Latent D...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Zhao Zhang Yulin Sun Zheng Zhang Yang Wang Guangcan Liu Meng Wang School of Computer Science and Technology Soochow University Suzhou China Key Laboratory of Knowledge Engineering with Big Data (Ministry of Education) Hefei University of Technology School of Computer Science and Information Engineering Hefei University of Technology Hefei China Bio-Computing Research Center Harbin Institute of Technology (Shenzhen) Shenzhen China School of Information and Control Nanjing University of Information Science and Technology Nanjing China

In this paper, we extend the popular dictionary pair learning (DPL) into the scenario of twin-projective latent flexible DPL under a structured twin-incoherence. Technically, a novel framework called Twin-Projective Latent Flexible DPL (TP-DPL) is proposed, which minimizes the twin-incoherence constrained flexibly-relaxed reconstruction error to avoid the possible over-fitting issue and produce accurate reconstruction. In this setting, TP-DPL integrates the twin-incoherence based latent flexible DPL and the joint embedding of codes as well as salient features by twin-projection into a unified model in an adaptive neighborhood-preserving manner. Therefore, TP-DPL can unify the procedures of salient feature representation and classification. The twin-incoherence constraint on coefficients and features can explicitly ensure high intra-class compactness and inter-class separation over them. TP-DPL also integrates the adaptive weighting to preserve local neighborhood of both coefficients and salient features within each class explicitly. For efficiency, TP-DPL selects the Frobenius-norm and abandons the costly l0/l1-norm for group sparse representation. Another byproduct is that TP-DPL can directly apply the class-specific twin-projective reconstruction residual to compute the label of data. Extensive results on public databases show that TP-DPL can deliver the state-of-the-art performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：