检索结果-内蒙古大学图书馆

41st International Conference on machine Learning, ICML 2024

作者： Di, Hao Ye, Haishan Chang, Xiangyu Dai, Guang Tsang, Ivor W. Center for Intelligent Decision-Making and Machine Learning School of Management Xi'an Jiaotong University China SGIT AI Lab State Grid Corporation of China China College of Computing and Data Science NTU Singapore Singapore

In decentralized optimization, m agents form a network and only communicate with their neighbors, which gives advantages in data ownership, privacy, and scalability. At the same time, decentralized stochastic gradient descent (SGD) methods, as popular decentralized algorithms for training large-scale machine learning models, have shown their superiority over centralized counterparts. Distributed stochastic gradient tracking (DSGT) (Pu & Nedić, 2021) has been recognized as the popular and state-of-the-art decentralized SGD method due to its proper theoretical guarantees. However, the theoretical analysis of DSGT (Koloskova et al., 2021) shows that its iteration complexity is (equation presented) where the doubly stochastic matrix W represents the network topology and CW is a parameter that depends on W. Thus, it indicates that the convergence property of DSGT is heavily affected by the topology of the communication network. To overcome the weakness of DSGT, we resort to the snapshot gradient tracking skill and propose two novel algorithms, snap-shot DSGT (SS DSGT) and accelerated snap-shot DSGT (ASS DSGT). We further justify that SS DSGT exhibits a lower iteration complexity compared to DSGT in the general communication network topology. Additionally, ASS DSGT matches DSGT's iteration complexity (equation presented) under the same conditions as DSGT. Numerical experiments validate SS DSGT's superior performance in the general communication network topology and exhibit better practical performance of ASS DSGT on the specified W compared to DSGT. Copyright 2024 by the author(s)

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

CONVOLUTIONAL OPTIMIZATION WITH CONVEX KERNEL AND POWER LIFT

arXiv

引用

arXiv 2025年

作者： Lu, Zhipeng Shenzhen MSU-BIT University Guangdong Laboratory of Machine Perception and Intelligent Computing Guangdong Shenzhen518172 China

We focus on establishing the foundational paradigm of a novel optimizationtheory based on convolution with convex kernels. Our goal is to devise amorally deterministic model of locating the global optima of an arbitraryfunction, which is distinguished from most commonly used statistical *** preliminary numerical results are provided to test the efficiency ofsome specific algorithms derived from our paradigm, which we hope to stimulatefurther practical interest. © 2025, CC BY.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Automatic Depression Detection Network Based on Facial Images and Facial Landmarks Feature Fusion

Automatic Depression Detection Network Based on Facial Image...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Hu, Min Xu, Lingxiang Liu, Lei Wang, Xiaohua Li, Hongbo Yang, Jiaoyun Hefei University of Technology Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine Hefei China Hefei University of Technology National Smart Eldercare International Science and Technology Cooperation Base Hefei China Hefei University of Technology Intelligent Interconnected Systems Laboratory of Anhui Province Hefei China

ISBN: (纸本)9798350386226

Artificial intelligence methods offer objectivity and convenience in automatic depression detection, however, current research often neglects the critical role of facial landmarks. This oversight results in insufficient spatial structure information and a lack of detailed local representation, which fails to capture the nuanced semantic information crucial for identifying depression-related clues. To address these issues, we introduce a novel dual-branch network model comprising the Landmark-Image-Landmark Net (LIL Net) and the Global Context Vision Transformer Net (GCVit Net). Through a dual-stream, multiscale, and cross-fusion strategy, LIL Net is designed to extract original facial image features alongside landmark features, prioritizing the detailed semantic information of potential depression clues. LIL Net employs an innovative LIL Attention approach to jointly learn multiscale features from facial landmarks and images, thereby enhancing the model's ability to capture fine-grained depression-related cues. Furthermore, the Multi-scale Feature Fusion (MSFF) module fuses the obtained multiscale features, augmenting the semantic expression of potential depression clues within facial landmarks via attention mechanisms. Meanwhile, the GCVit Net branch network supplements global information by extracting global facial features. Finally, the features from both branches are concatenated to enhance the accuracy of depression degree predictions. Experimental results demonstrate that our model has superior performance in detecting depression compared to existing methods. We release our code at https://***/xlx777/LIL-Net. © 2024 IEEE.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

EKIGCN: External Knowledge Injected Graph Convolutional Networks for Aspect-based Sentiment Analysis 9

EKIGCN: External Knowledge Injected Graph Convolutional Netw...

引用

9th IEEE International Conference on Cloud computing and Intelligence Systems, CCIS 2023

作者： Shi, Xuefeng Hu, Min Ren, Fuji Shi, Piao Hefei University of Technology Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine Hefei230000 China University of Electronic Science and Technology of China Sichuan Chengdu611731 China

ISBN: (纸本)9798350304428

Aspect-based sentiment analysis (ABSA) is a challenging subtask in the natural language processing community, which aims to determine the sentiment polarity about the corresponding specific aspect terms. In existing works, external knowledge has been proven effective to improve the ABSA tasks' performance. However, it is still essential to study further to find a universal method for ABSA tasks. Therefore, we propose an external knowledge-injected graph convolutional network (EKIGCN) to enhance the sentiment representation with commonsense. Specifically, Sentic Net is leveraged to provide external knowledge to construct an affective dependency graph. In order to fully exploit the effectiveness of affective knowledge, a shallow mutual interaction is utilized to fuse the learned contextual representation and the graph. Besides, to make EKIGCN sensitive to aspect terms, an aspect-aware multi-head attention module is also employed to enhance the corresponding relations between aspect and contextual words. Extensive experiments are conducted on three popular datasets to validate the effectiveness of our proposed model, and the results outperform the mentioned state-of-the-art methods in ABSA tasks. © 2023 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks 38

Rethinking the Membrane Dynamics and Optimization Objectives...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Shen, Hangchi Zheng, Qian Wang, Huamin Pan, Gang College of Artificial Intelligence Southwest University China Chongqing Key Laboratory of Brain Inspired Computing and Intelligent Chips China The State Key Lab of Brain-Machine Intelligence Zhejiang University China College of Computer Science and Technology Zhejiang University China

Despite spiking neural networks (SNNs) have demonstrated notable energy efficiency across various fields, the limited firing patterns of spiking neurons within fixed time steps restrict the expression of information, which impedes further improvement of SNN performance. In addition, current implementations of SNNs typically consider the firing rate or average membrane potential of the last layer as the output, lacking exploration of other possibilities. In this paper, we identify that the limited spike patterns of spiking neurons stem from the initial membrane potential (IMP), which is set to 0. By adjusting the IMP, the spiking neurons can generate additional firing patterns and pattern mappings. Furthermore, we find that in static tasks, the accuracy of SNNs at each time step increases as the membrane potential evolves from zero. This observation inspires us to propose a learnable IMP, which can accelerate the evolution of membrane potential and enables higher performance within a limited number of time steps. Additionally, we introduce the last time step (LTS) approach to accelerate convergence in static tasks, and we propose a label smooth temporal efficient training (TET) loss to mitigate the conflicts between optimization objective and regularization term in the vanilla TET. Our methods improve the accuracy by 4.05% on ImageNet compared to baseline and achieve state-of-the-art performance of 87.80% on CIFAR10-DVS and 87.86% on N-Caltech101. The code is available at https://***/StephenTaylor1998/IMP-SNN. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Octree K-nearest neighbor parallel query 4

Octree K-nearest neighbor parallel query

引用

4th International Conference on Computer Vision and Pattern Analysis, ICCPA 2024

作者： Nong, Jian Xu, Jian He, Xi Li, Wenge Huang, Hongben Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Wuzhou543002 China High Performance Computing Laboratory Wuzhou University Wuzhou543002 China CCTEG Taiyuan Research Institute Co. Ltd. Taiyuan030006 China

ISBN: (纸本)9781510682528

Nearest neighbor queries have widespread applications in fields such as computer vision, data mining, pattern recognition, document retrieval, and geographic information systems. The traditional k-nearest neighbor algorithm based on branch-and-bound and octree requires the use of a priority queue to maintain the result set. However, in large-scale data scenarios, the efficiency of the priority queue decreases, leading to a decline in query efficiency. Additionally, implementing parallelism using tree-based storage structures, such as octree, can be challenging. This paper proposes a parallel k-nearest neighbor query algorithm for multi-core CPU environments based on octree. This method adopts a divide-and-conquer approach by dividing the dataset into multiple parts and constructing octree for each part. Then, it utilizes parallel max-heaps to calculate the distances of pruned data and selects the nearest K elements to achieve K-nearest neighbor query. This method utilizes multiprocessors for tree construction and result set maintenance, further accelerating the query speed. Experimental results show that in large-scale data scenarios, this method achieves significant acceleration compared to sequential query. © 2024 SPIE.

关键词： Nearest neighbor search

来源：评论

学校读者我要写书评

暂无评论

Decoding Illusion Perception: A Comparative Analysis of Deep Neural Networks in the Müller-Lyer Illusion

Decoding Illusion Perception: A Comparative Analysis of Deep...

引用

2023 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2023

作者： Zhang, Hongtao Yoshida, Shinichi Li, Zhen Graduate School of Engineering Kochi University of Technology Kochi Japan School of Information Kochi University of Technology Kochi Japan Shenzhen MSU-BIT University Guangdong Laboratory of Machine Perception and Intelligent Computing Shenzhen China

ISBN: (纸本)9798350337020

Visual illusions, as powerful tools for revealing the strategies and limitations of brain visual processing, have garnered attention in neural network research. This study aims to investigate the performance of various deep neural networks (DNNs) in handling the Müller-Lyer illusion task, thereby providing insights into the mechanisms of the brain's visual system in processing visual illusions. We adopted two types of visual stimuli and recruited 12 participants for a length-matching task. Based on the participants' perceptual data, we employed multiple classical neural network models to test their responses to each type of stimuli. To analyze the differences in model performance, we employed representational dissimilarity matrices (RDM) analysis and utilized Grad-CAM techniques to visualize neural network activation maps. The results revealed that some neural networks could simulate human visual illusion phenomena but exhibit differences in representational similarity and feature preferences. These findings provide a preliminary understanding of the computational principles and mechanisms of the brain's visual system and offer insights for future research in building more robust and efficient neural network models. © 2023 IEEE.

关键词： computer vision deep learning neural networks optical illusion visual system

来源：评论

学校读者我要写书评

暂无评论

On Interval Perturbation of the α-Universal Triple I Algorithm for Unified Interval Implications 1

引用

3rd International Conference on Artificial Intelligence Logic and Applications, AILA 2023

作者： Tang, Yiming Huang, Yifan Qi, Zhiyuan Zhang, Li Gao, Jianwei Anhui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine School of Computer and Information Hefei University of Technology Hefei230601 China College of Overseas Education Nanjing Tech University Nanjing211816 China

ISBN: (数字)9789819978694

ISBN: (纸本)9789819978687

In the fuzzy field, Zadeh put forward the CRI (compositional rule of inference) algorithm to deal with fuzzy reasoning. Later, Wang put forward the triple I algorithm, which is logically more complete than CRI. Thenceforward, Tang found that the harsh conditions of the triple identical implication operators lead to low system performance, so the α-universal triple I (α-UTI) algorithm is proposed. In actual applications, the quality of fuzzy inference algorithm directly is determined by the robustness of inference algorithm. This study researches the interval robustness of the α-UTI algorithm under the interval-valued environment. The interval perturbation performance of the α-UTI algorithm is respectively revealed for the R-interval implication, the triangular modulus interval implication, the S-interval implication and the QL-interval implication. It is observed that the α-UTI algorithm has good interval perturbation performance for single rule or multi-rules, in which unified interval implications are employed. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Fuzzy inference

来源：评论

学校读者我要写书评

暂无评论

DMHFR:Decoder with Multi-Head Feature Receptors for Tract Image Segmentation

引用

Computers, Materials & Continua 2025年第3期82卷 4841-4862页

作者： Jianuo Huang Bohan Lai Weiye Qiu Caixu Xu Jie He Department of Endoscopy Center Zhongshan Hospital(Xiamen)Fudan UniversityXiamen361015China School of Computing and Data Science Xiamen University MalaysiaSepang43900Malaysia School of Computer Science and Techonology Tongji UniversityShanghai200092China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou UniversityWuzhou543002China Xiamen Clinical Research Center for Cancer Therapy Xiamen361015China

The self-attention mechanism of Transformers,which captures long-range contextual information,has demonstrated significant potential in image ***,their ability to learn local,contextual relationships between pixels requires further *** methods face challenges in efficiently managing multi-scale fea-tures of different granularities from the encoder backbone,leaving room for improvement in their global representation and feature extraction *** address these challenges,we propose a novel Decoder with Multi-Head Feature Receptors(DMHFR),which receives multi-scale features from the encoder backbone and organizes them into three feature groups with different granularities:coarse,fine-grained,and full *** groups are subsequently processed by Multi-Head Feature Receptors(MHFRs)after feature capture and modeling *** include two Three-Head Feature Receptors(THFRs)and one Four-Head Feature Receptor(FHFR).Each group of features is passed through these MHFRs and then fed into axial transformers,which help the model capture long-range dependencies within the *** three MHFRs produce three distinct feature *** output from the FHFR serves as auxiliary auxiliary features in the prediction head,and the prediction output and their losses will eventually be *** results show that the Transformer using DMHFR outperforms 15 state of the arts(SOTA)methods on five public ***,it achieved significant improvements in mean DICE scores over the classic Parallel Reverse Attention Network(PraNet)method,with gains of 4.1%,2.2%,1.4%,8.9%,and 16.3%on the CVC-ClinicDB,Kvasir-SEG,CVC-T,CVC-ColonDB,and ETIS-LaribPolypDB datasets,respectively.

关键词： Medical image segmentation feature exploration feature aggregation deep learning multi-head feature receptor

来源：评论

学校读者我要写书评

暂无评论

Large-Scale Riemannian Meta-Optimization via Subspace Adaptation

arXiv

引用

arXiv 2025年

作者： Yu, Peilin Wu, Yuwei Gao, Zhi Fan, Xiaomeng Jia, Yunde Beijing Key Laboratory of Intelligent Information Technology School of Computer Science & Technology Beijing Institute of Technology China Guangdong Laboratory of Machine Perception and Intelligent Computing Shenzhen MSU-BIT University China

Riemannian meta-optimization provides a promising approach to solving non-linear constrained optimization problems, which trains neural networks as optimizers to perform optimization on Riemannian manifolds. However, existing Riemannian meta-optimization methods take up huge memory footprints in large-scale optimization settings, as the learned optimizer can only adapt gradients of a fixed size and thus cannot be shared across different Riemannian parameters. In this paper, we propose an efficient Riemannian meta-optimization method that significantly reduces the memory burden for large-scale optimization via a subspace adaptation scheme. Our method trains neural networks to individually adapt the row and column subspaces of Riemannian gradients, instead of directly adapting the full gradient matrices in existing Riemannian meta-optimization methods. In this case, our learned optimizer can be shared across Riemannian parameters with different sizes. Our method reduces the model memory consumption by six orders of magnitude when optimizing an orthogonal mainstream deep neural network (e.g. ResNet50). Experiments on multiple Riemannian tasks show that our method can not only reduce the memory consumption but also improve the performance of Riemannian meta-optimization. © 2025, CC BY-NC-ND.

关键词： Constrained optimization

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：