检索结果-内蒙古大学图书馆

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Weixin An Yuanyuan Liu Fanhua Shang Hongying Liu Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education School of Artificial Intelligence Xidian University China College of Intelligence and Computing Tianjin University China Medical School Tianjin University China and Peng Cheng Lab Shenzhen China

ISBN: (纸本)9798331314385

Many zeroth-order (ZO) optimization algorithms have been developed to solve nonconvex minimax problems in machine learning and computer vision areas. However, existing ZO minimax algorithms have high complexity and rely on some strict restrictive conditions for ZO estimations. To address these issues, we design a new unified ZO gradient descent extragradient ascent (ZO-GDEGA) algorithm, which reduces the overall complexity to O(dε-6) to find an ε-stationary point of the function ψ for nonconvex-concave (NC-C) problems, where d is the variable dimension. To the best of our knowledge, ZO-GDEGA is the first ZO algorithm with complexity guarantees to solve stochastic NC-C problems. Moreover, ZO-GDEGA requires weaker conditions on the ZO estimations and achieves more robust theoretical results. As a by-product, ZO-GDEGA has advantages on the condition number for the NC-strongly concave case. Experimentally, ZO-GDEGA can generate more effective poisoning attack data with an average accuracy reduction of 5%. The improved AUC performance also verifies the robustness of gradient estimations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fast Node Selection of Networked Radar Based on Transfer Reinforcement Learning 5th

Fast Node Selection of Networked Radar Based on Transfer R...

引用

5th IFIP TC 12 International Conference on Intelligence Science, ICIS 2022

作者： Cao, Yanjun Wang, Yuan Guo, Jingjing Han, Li Zhang, Chao Zhu, Jin Zhang, Tianyang Zhang, Xiangrong The 54th Research Institute of China Electronics Technology Group Corporation Beijing China Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Xidian University Shaanxi Xian710071 China National Laboratory of Radar Signal Processing Xidian University Xian710071 China

ISBN: (纸本)9783031149023

The networked radar system can synthesize different echo signals received by various radars and realize the cooperative detection of multiple radars, becoming more and more critical for data fusion sharing and network collaboration. However, due to the large number and wide range of nodes in the networked radar system, there exists a redundancy problem in radar node assignment, which causes additional resource consumption and slows down the task execution speed of radar node selection. To solve the above problem, this paper proposes a fast radar node selection method based on transfer reinforcement learning to quickly select the optimal and minimum node resources. The proposed method devises a novel reward function for the Monte Carlo Tree and a different termination state of iteration to select the minimize the number of radar nodes. In order to further accelerate the selection of radar nodes, transfer reinforcement learning is presented to fully leverage the previous knowledge. Experimental results show that our proposed method can quickly select the optimal and minimum radar nodes in a brief period, significantly improving the speed of radar node selection in the networked radar. © 2022, IFIP International Federation for Information Processing.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Self-Constrained Graph Stochastic Neural Networks for Graphstructure Learning

SSRN

引用

SSRN 2022年

作者： Jia, Nan Tian, Xiaolin Wang, Kai Huang, Xiaocui Key Lab of Intelligent Perception and Image Understanding of Ministry of Education International Center of Intelligent Perception and Computation Xidian University Xi’an710071 China

Graph Neural Networks (GNNs) have emerged as a powerful and flexible framework for graph structure ***, most existing GNNs heavily rely on graph structure and may suffer from the limitation of over-smoothingwhen we enlarge the feature propagation steps. In this paper, a simple yet effective framework—the Self-ConstrainedGraph Stochastic Neural Networks (SCGSNN) is proposed to address these two issues. Firstly, we use the multibranchgenerating structure to learn the adjacency matrix and the GNN parameters simultaneously in SCGSNN,aiming to obtain multiple different first-order adjacency matrices. Then, we develop the self-constrained mechanismto constrain each generation structure branch to infer a more realistic graph structure. Finally, we apply the mixedmulti-level information aggregation to alleviate the issue of over-smoothing. Experimental results show that SCGSNNcan be extended to large graphs with million-level nodes and apparently outperforms state-of-the-art GNNs baselinesfor the node classification task. © 2022, The Authors. All rights reserved.

关键词： Graphic methods

来源：评论

学校读者我要写书评

暂无评论

Hyper-parameter Tuning of Federated Learning Based on Particle Swarm Optimization 7

Hyper-parameter Tuning of Federated Learning Based on Partic...

引用

7th IEEE International Conference on Cloud Computing and Intelligence Systems, CCIS 2021

作者： Li, Zhiyuan Li, Hao Zhang, Mingyang Xidian University School of Electronic Engineering Key Laboratory of lntelligent Perception and Image Understanding of Ministry of Education of China Xi'an710071 China

ISBN: (纸本)9781665441490

The learning task of federated learning (FL) is solved by a federation of a center server and individual clients. Contrary to traditional deep learning models, federated learning consists of two parts, the global model and individual models. However, since the process is carried out by two parts of equal importance, it is ideal to deal with both parts simultaneously. To achieve superior performance, federated learning requires carefully selected hyper-parameters, which has more hyper-parameters than those of traditional deep learning models. To solve this tuning problem, we propose a method using particle swarm optimization (PSO) algorithm to tune the hyperparameters of federated learning. PSO algorithm is a gradient-free, stochastic optimization method which is better than grid search method when it comes to large search space. It helps locate the optimal combination of multiple parameters. In this article, we focus on applying PSO method on tuning the hyper-parameters of FL models, and prove that it is an efficient way to acquire satisfactory results. Experiments on MNIST dataset with convolution neural networks have proved the superiority of the proposed method. © 2021 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Integrating Language-image Prior into EEG Decoding for Cross-Task Zero-Calibration RSVP-BCI

arXiv

引用

arXiv 2025年

作者： Li, Xujin Wei, Wei Qiu, Shuang Zhang, Xinyi Li, Fu He, Huiguang The Key Laboratory of Brain Cognition and Brain-inspired Intelligence Technology Institute of Automation Chinese Academy of Sciences Beijing100190 China Beijing100049 China The Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education The School of Artificial Intelligence Xidian University Xi’an710071 China

Rapid Serial Visual Presentation (RSVP)-based Brain-Computer Interface (BCI) is an effective technology used for information detection by detecting Event-Related Potentials (ERPs). The current RSVP decoding methods can perform well in decoding EEG signals within a single RSVP task, but their decoding performance significantly decreases when directly applied to different RSVP tasks without calibration data from the new tasks. This limits the rapid and efficient deployment of RSVP-BCI systems for detecting different categories of targets in various scenarios. To overcome this limitation, this study aims to enhance the cross-task zero-calibration RSVP decoding performance. First, we design three distinct RSVP tasks for target image retrieval and build an open-source dataset containing EEG signals and corresponding stimulus images. Then we propose an EEG with Language-image Prior fusion Transformer (ELIPformer) for cross-task zero-calibration RSVP decoding. Specifically, we propose a prompt encoder based on the language-image pre-trained model to extract language-image features from task-specific prompts and stimulus images as prior knowledge for enhancing EEG decoding. A cross bidirectional attention mechanism is also adopted to facilitate the effective feature fusion and alignment between the EEG and language-image features. Extensive experiments demonstrate that the proposed model achieves superior performance in cross-task zero-calibration RSVP decoding, which promotes the RSVP-BCI system from research to practical application. Copyright © 2025, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Transformers Meet Visual Learning understanding: A Comprehensive Review

arXiv

引用

arXiv 2022年

作者： Yang, Yuting Jiao, Licheng Liu, Xu Liu, Fang Yang, Shuyuan Feng, Zhixi Tang, Xu The Key Laboratory of Intelligent Perception and Image Understanding The Ministry of Education of China International Research Center of Intelligent Perception and Computation School of Artificial Intelligence Xidian University Xi’an China

Dynamic attention mechanism and global modeling ability make Transformer show strong feature learning ability. In recent years, Transformer has become comparable to CNNs methods in computer vision. This review mainly investigates the current research progress of Transformer in image and video applications, which makes a comprehensive overview of Transformer in visual learning understanding. First, the attention mechanism is reviewed, which plays an essential part in Transformer. And then, the visual Transformer model and the principle of each module are introduced. Thirdly, the existing Transformer-based models are investigated, and their performance is compared in visual learning understanding applications. Three image tasks and two video tasks of computer vision are investigated. The former mainly includes image classification, object detection, and image segmentation. The latter contains object tracking and video classification. It is significant for comparing different models’ performance in various tasks on several public benchmark data sets. Finally, ten general problems are summarized, and the developing prospects of the visual Transformer are given in this review. © 2022, CC BY.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Meta-Feature Relearning with Embedded Label Semantics and Reweighting for Few-Shot Object Detection

引用

Jisuanji Xuebao/Chinese Journal of Computers 2022年第12期45卷 2561-2575页

作者： Li, Peng-Fang Liu, Fang Li, Ling-Ling Liu, Xu Feng, Zhi-Xi Jiao, Li-Cheng Xiong, Yi-Meng School of Artificial Intelligent Xidian University Xi'an710071 China Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Xi'an710071 China International Research Center for Intelligent Perception and Computation Xi'an710071 China Joint International Research Laboratory of Intelligent Perception and Computation Xi'an710071 China

Object Detection methods based on Deep Learning (DL) have been able to achieve good detection accuracy. Nevertheless, DL, as a data-driven technique, relies heavily on massive labeled data. Currently, Few-Shot Learning has been extensively studied, as it can alleviate the reliance on a large number of labeled samples. In this paper, we focus on the research on Few-Shot Object Detection (FSOD). In FSOD, the deep model is first learned on the base class data that must have enough labeled samples. The model then continues to learn new classes with only a few labeled samples. The ultimate goal of model learning is to quickly adapt to the identification and localization of new classes of objects. In FSOD, on the one hand, compared with the base class, the new class has few samples available. On the other hand, the new class and the base class contain different target classes. This results in that FSOD methods are generally able to learn better base class features, but the learned new class features are weakly separable. To enhance the separability of meta-features for new classes, in this paper, we propose a Meta-Feature Relearning with Embedded Label Semantics and Reweighting Few-Shot Object Detection method. The proposed method can transfer the new class-related features from the base class to the new class, thereby enhancing the meta-features of the new class. In detail, in the few-shot training stage, firstly, we construct a word vector label semantic graph generation module. The word vector label semantic graph generation module introduces the label semantic information to generate the word vector label semantic graph. The generated word vector label semantic graph is used to model the semantic association between the base class and the new class. Meanwhile, we construct a label semantic embedding module. The label semantic embedding module first integrates the semantic association between the base class and the new class. Then, the meta-features of the support set

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Pixel-to-Abundance Translation: Conditional Generative Adversarial Networks Based on Patch Transformer for Hyperspectral Unmixing

arXiv

引用

arXiv 2023年

作者： Wang, Li Zhang, Xiaohua Li, Longfei Meng, Hongyun Cao, Xianghai School of Artificial Intelligence Xidian University Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Xi’an710071 China School of Mathematics and Statistics Xidian University Xi’an710071 China

Spectral unmixing is a significant challenge in hyperspectral image processing. Existing unmixing methods utilize prior knowledge about the abundance distribution to solve the regularization optimization problem, where the difficulty lies in choosing appropriate prior knowledge and solving the complex regularization optimization problem. To solve these problems, we propose a hyperspectral conditional generative adversarial network (HyperGAN) method as a generic unmixing framework, based on the following assumption: the unmixing process from pixel to abundance can be regarded as a transformation of two modalities with an internal specific relationship. The proposed HyperGAN is composed of a generator and discriminator, the former completes the modal conversion from mixed hyperspectral pixel patch to the abundance of corresponding endmember of the central pixel and the latter is used to distinguish whether the distribution and structure of generated abundance are the same as the true ones. We propose hyperspectral image (HSI) Patch Transformer as the main component of the generator, which utilize adaptive attention score to capture the internal pixels correlation of the HSI patch and leverage the spatial-spectral information in a fine-grained way to achieve optimization of the unmixing process. Experiments on synthetic data and real hyperspectral data achieve impressive results compared to state-of-the-art competitors. Copyright © 2023, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Global Spatio-Temporal Fusion-based Traffic Prediction Algorithm with Anomaly Aware

Global Spatio-Temporal Fusion-based Traffic Prediction Algor...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Chaoqun Liu Xuanpeng Li Chen Gong Guangyu Li PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education and Jiangsu Key Lab of Image and Video Understanding for Social Security Nanjing University of Science and Technology Southeast University Nanjing China

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

Traffic prediction is an indispensable component of urban planning and traffic management. Achieving accurate traffic prediction hinges on the ability to capture the potential spatio-temporal relationships among road sensors. However, the majority of existing works focus on local short-term spatiotemporal correlations, failing to fully consider the interactions of different sensors in the long-term state. In addition, these works do not analyze the influences of anomalous factors, or have insufficient ability to extract personalized features of anomalous factors, which make them ineffectively capture their spatiotemporal influences on traffic prediction. To address the aforementioned issues, We propose a global spatio-temporal fusionbased traffic prediction algorithm that incorporates anomaly awareness. Initially, based on the designed anomaly detection network, we construct an efficient anomalous factors impacting module (AFIM), to evaluate the spatio-temporal impact of unexpected external events on traffic prediction. Furthermore, we propose a multi-scale spatio-temporal feature fusion module (MTSFFL) based on the transformer architecture, to obtain all possible both long and short term correlations among different sensors in a wide-area traffic environment for accurate prediction of traffic flow. Finally, experiments are implemented based on real-scenario public transportation datasets (PEMS04 and PEMS08) to demonstrate that our approach can achieve stateof-the-art performance.

关键词： Correlation Accuracy Roads Urban planning Sensor fusion Prediction algorithms Feature extraction Transformers Sensors Spatiotemporal phenomena

来源：评论

学校读者我要写书评

暂无评论

Multi-Task Multi-Scale Feature Selection for Point Cloud Registration

引用

IEEE Transactions on Evolutionary Computation 2025年

作者： Wu, Yue Luo, Chuang Gong, Maoguo Ding, Hangqi Sheng, Jinlong Miao, Qiguang Li, Hao Ma, Wenping He, Hao Xidian University School of Computer Science and Technology Xi'an710071 China Xidian University School of Electronic Engineering Key Laboratory of Collaborative Intelligence Systems Ministry of Education Xi'an710071 China Inner Mongolia Normal University College of Mathematics Science Hohhot010028 China Xidian University School of Artificial Intelligence Key Laboratory of Intelligent Perception and Image Understanding Ministry of Education Xi'an710071 China

3D point cloud registration is a process of solving the geometric transformation between two point clouds. This process is an important issue in computer vision and pattern recognition. The registration methods based on geometric features are highly sensitive to the scale of feature extraction. Changes in scale can introduce inaccuracies in feature descriptions, thereby compromising the reliability of the registration results. To mitigate the impact of feature scale on the outcomes and the high-dimensional issue arising from features of different scales, we propose a method for multi-scale point cloud feature selection. We solve the high-dimensional problem of feature selection by designing a multi-task framework. By designing a mutual information dimensionality reduction method, we decomposed the high-dimensional feature selection task of different descriptors with multi-scale features into multiple related low-dimensional feature selection tasks. Then, by means of the knowledge transfer among these low-dimensional feature selection tasks, we sought the best feature subset to obtain more robust feature information. We evaluate the effectiveness of our method by conducting extensive experiments on various datasets. The experimental results show that the method outperforms other feature descriptors in terms of descriptive power and robustness and improves the effectiveness of point cloud registration. © 1997-2012 IEEE.

关键词： Feature Selection

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：