检索结果-内蒙古大学图书馆

13th International Conference on intelligent control and Information processing, ICICIP 2025

作者： Li, Jiahao Wang, Zhongshuai Ghazali, Kamarul Hawari Bin Yan, Suqing Lan, Rushi Sun, Xiyan Luo, Xiaonan Guangxi Key Lab. of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin541004 China Centre for Advanced Industrial Technology University of Malaysia Pahang Al-Sultan Abdullah Pahang Pekan26600 Malaysia Int. Joint Research Lab. of Spatio-temporal Information and Intelligent Location Services Guilin University of Electronic Technology Guilin541004 China

ISBN: (纸本)9798331516147

Privacy-preserving image generation is particularly crucial in fields like healthcare, where data are both sensitive and limited. However, effective privacy preservation often compromises the visual quality and utility of the generated images due to privacy budget constraints. To address this issue, in this paper, We propose a novel network architecture, IRSEnet, which combines multi-scale feature extraction technology and residual channel attention mechanisms, aiming to enhance the visual quality of generated images and improve the performance of downstream classification tasks under differential privacy. The differential privacy mechanism ensures the security of sensitive data during training, while the multi-scale feature extraction module enhances feature extraction capabilities through parallel convolutional layers at multiple scales. Additionally, the channel attention module dynamically adjusts channel weights to focus on the most discriminative features. Experimental results demonstrate that this model significantly improves the utility of generated images and the accuracy of downstream classification tasks while preserving privacy. Future work will explore the application of this approach on larger datasets and across more diverse tasks. © 2025 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Output-Feedback Global Consensus of Discrete-Time Multiagent Systems Subject to Input Saturation via Q-Learning Method

引用

IEEE TRANSACTIONS ON CYBERNETICS 2022年第3期52卷 1661-1670页

作者： Long, Mingkang Su, Housheng Zeng, Zhigang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Educ Minist China Wuhan 430074 Peoples R China

This article proposes a Q-learning (QL)-based algorithm for global consensus of saturated discrete-time multiagent systems (DTMASs) via output feedback. According to the low-gain feedback (LGF) theory, control inputs of the saturated DTMASs can avoid the saturation by utilizing the control policies with LGF matrices, which were computed from the modified algebraic Riccati equation (MARE) by requiring the information of system dynamics in most previous works. However, in this article, we first find the lower bound on the real part of Laplacian matrices' nonzero eigenvalues of directed network topologies. Then, we define a test control input and propose a Q-function to derive a QL Bellman equation, which plays an essential part of the QL algorithm. Subsequently, different from the previous works, the output-feedback gain (OFG) matrix of this article can be obtained by limited iterations of the QL algorithm without requiring the information of agent dynamics and network topologies of the saturated DTMASs. Furthermore, the saturated DTMASs can achieve global consensus rather than the semiglobal consensus of the previous results. Finally, the effectiveness of the QL algorithm is confirmed via two simulations.

关键词： Discrete time global consensus input saturation multiagent systems output feedback Q-learning (QL) algorithm

来源：评论

学校读者我要写书评

暂无评论

Global Exponential Stability of Impulsive Delayed Neural Networks on Time Scales Based on Convex Combination Method

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2022年第5期52卷 3015-3024页

作者： Wan, Peng Zeng, Zhigang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Educ Minist China Wuhan 430074 Peoples R China

The published stability criteria for impulsive neural networks are scale-free on time line, which is only appropriate for discrete or continuous ones. The issue of global exponential stability for impulsive delayed neural networks on time scales is analyzed by employing the convex combination method in this article. Several algebraic and linear matrix inequality conditions are proved by constructing impulse-dependent Lyapunov functionals and using timescale inequality techniques. Unlike the published works, impulsive control strategies can be designed by utilizing our theoretical results to stabilize delayed neural networks on time scales if they are unstable before introducing impulses. Sufficient criteria for global exponential stability in this article are derived based on the timescale theory, and they are applicable to discrete-time impulsive neural networks, their continuous-time analogues, and neural networks whose states are discrete at one time and continuous at another time. Four numerical examples are offered to demonstrate the effectiveness and superiority of our new theoretical results in the end.

关键词： Biological neural networks Stability criteria control theory Synchronization Technological innovation Statistics Sociology Delay exponential stability impulse neural networks time scale

来源：评论

学校读者我要写书评

暂无评论

Event-triggered H∞ Consensus of Leader-follower Multi-agent Systems 43

Event-triggered H∞ Consensus of Leader-follower Multi-agent...

引用

43rd Chinese control Conference, CCC 2024

作者： Gong, Cheng Ai, Chenguang Zuo, Xuemin Ma, Zhuang Dai, W. Hebei Jianzhi Casting Group Ltd Pangzhuangzi Yutian County Tangshan064100 China College Tangshan University Key Lab of Intelligent Data Information Processing and Control of Hebei Province Key Lab of Intelligent Motion Control System of Tangshan City Intelligence and Information Engineering Tangshan063000 China North China University of Science and Technology College of Electrical Engineering Tangshan063210 China Kernel University FullertonCA92832 United States

ISBN: (纸本)9789887581581

This paper discusses the H∞ consensus problem of leader-follower multi-agent systems. The controller for each agent is crafted to utilize comprehensive information from all connected agents, while an innovative event-triggered strategy is incorporated to minimize communication costs. Based on the Lyapunov function theory, an H∞ controller is proposed that can ensure both system stability and a prescribed disturbance attenuation level γ. Finally, a numerical example validated the effectiveness of the proposed method. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Refine for Semantic Segmentation Based on Parallel Convolutional Network with Attention Model

引用

NEURAL processing LETTERS 2021年第6期53卷 4177-4188页

作者： Peng, Gang Yang, Shiqi Wang, Hao Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Minist Educ Key Lab Image Proc & Intelligent Control Wuhan Peoples R China

High-precision semantic segmentation methods require global information and more detailed local features. It is difficult for ordinary convolutional neural networks to efficiently use this information. In response to the above issues, this paper uses the attention to scale method and proposes a novel attention model for semantic segmentation, which aggregates multi-scale and context features to refine prediction. Specifically, the skeleton convolutional neural network framework takes in multiple different scales inputs, by which means the CNN can get representations in different scales. The proposed attention model will handle the features from different scale streams respectively and integrate them. Then location attention branch of the model learns to softly weight the multi-scale features at each pixel location. Moreover, we add an recalibrating branch, parallel to where location attention comes out, to recalibrate the score map per class. We achieve quite competitive results on PASCAL VOC 2012 and ADE20K datasets, which surpass baseline and related works.

关键词： Semantic segmentation Parallel convolutional network Attention model Multi-scale Multi-dilation

来源：评论

学校读者我要写书评

暂无评论

Late Better than Early: A Decision-Level Information Fusion Approach for Rgb-Thermal Crowd Counting with Illumination Awareness

SSRN

引用

SSRN 2024年

作者： Cheng, Jian Feng, Chen Xiao, Yang Cao, Zhiguo Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

In this paper, we make the first research effort to address the RGB-Thermal (RGB-T) crowd counting problem with decision-level late fusion manner. Being different from the existing pixel-level or feature-level fusion methods, our proposition chooses to fuse the density maps yielded by RGB and thermal counterparts via spatially adaptive weighting with RGB illumination-aware attention. Our key intuition to conduct RGB-T density map fusion lies in 3 main folders. First, compared with the raw RGB-T images or convolutional feature maps, RGB-T density maps contain stronger counting-wise semantic meanings. Secondly, they are also of high spatial resolution for revealing fine local details. Thirdly, RGB-T density maps are complementary to each other for effective counting pattern characterization. To fuse them adaptively, a spatial weighting map is generated concerning the issues of RGB illumination awareness and local counting pattern characterization ability jointly. To the best of our knowledge, we are the first to leverage RGB-T crowd counting via concerning these 2 issues in a unified way. Meanwhile, cross-modality feature interaction is conducted between RGB and thermal modalities to facilitate spatial weighting map generation. The experiments on 2 well-established RGB-T crowd counting datasets, i.e., RGBT-CC and DroneRGBT) verify the superiority of our proposition. The source code and pretrained models will be released upon acceptance at https://***/hustaia/DLF-IA. © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

A Comparison of Large-Scale MOEAs with Informed Initialization for Voltage Transformer Ratio Error Estimation 1

引用

17th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2022

作者： Li, Lianghao He, Cheng Li, Hongbin Key Laboratory of Image Information Processing and Intelligent Control of Education Ministry of China School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China School of Electrical and Electronic Engineering Huazhong University of Science and Technology Wuhan430074 China

ISBN: (数字)9789819915491

ISBN: (纸本)9789819915484

Large-scale multiobjective optimization problems (LSMOPs) exist widely in real-world applications. The large number of decision variables in LSMOP leads to a tremendous high-dimensional search space, which is still challenging for existing multiobjective evolutionary algorithms (MOEAs). The voltage transformer ratio error estimation (TREE) problem is a typical LSMOP in real-world applications. A number of large-scale multiobjective evolutionary algorithms (LSMOEAs) have been adopted for solving the TREE problem, but so far there is still considerable potential for improvement in their results. In this paper, we propose a informed initialization for solving the TREE problem. The physical properties of the TREE problem are used to construct an initial set of solutions with specific characteristics. The proposed initialization method is tested against the default initialization method on five TREE problems by being applied on six state-of-the-art LSMOEAs and a classical MOEA. The experimental results demonstrate that the initialization method has a significant effect on the performance of LSMOEAs in solving the TREE problems. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Cross-Task Inconsistency Based Active Learning (CTIAL) for Emotion Recognition

arXiv

引用

arXiv 2024年

作者： Xu, Yifan Jiang, Xue Wu, Dongrui Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

Emotion recognition is a critical component of affective computing. Training accurate machine learning models for emotion recognition typically requires a large amount of labeled data. Due to the subtleness and complexity of emotions, multiple evaluators are usually needed for each affective sample to obtain its ground-truth label, which is expensive. To save the labeling cost, this paper proposes an inconsistency-based active learning approach for cross-task transfer between emotion classification and estimation. Affective norms are utilized as prior knowledge to connect the label spaces of categorical and dimensional emotions. Then, the prediction inconsistency on the two tasks for the unlabeled samples is used to guide sample selection in active learning for the target task. Experiments on within-corpus and cross-corpus transfers demonstrated that cross-task inconsistency could be a very valuable metric in active learning. To our knowledge, this is the first work that utilizes prior knowledge on affective norms and data in a different task to facilitate active learning for a new task, even the two tasks are from different datasets. © 2024, CC BY.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

Knowledge-Data Fusion Based Source-Free Semi-Supervised Domain Adaptation for Seizure Subtype Classification

arXiv

引用

arXiv 2024年

作者： Peng, Ruimin An, Jiayu Wu, Dongrui Key Laboratory of the Ministry of Education for Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

Electroencephalogram (EEG)-based seizure subtype classification enhances clinical diagnosis efficiency. Source-free semi-supervised domain adaptation (SF-SSDA), which transfers a pre-trained model to a new dataset with no source data and limited labeled target data, can be used for privacy-preserving seizure subtype classification. This paper considers two challenges in SF-SSDA for EEG-based seizure subtype classification: 1) How to effectively fuse both raw EEG data and expert knowledge in classifier design? 2) How to align the source and target domain distributions for SF-SSDA? We propose a Knowledge-Data Fusion based SF-SSDA approach, KDF-MutualSHOT, for EEG-based seizure subtype classification. In source model training, KDF uses Jensen-Shannon Divergence to facilitate mutual learning between a feature-driven Decision Tree-based model and a data-driven Transformer-based model. To adapt KDF to a new target dataset, an SF-SSDA algorithm, MutualSHOT, is developed, which features a consistency-based pseudo-label selection strategy. Experiments on the public TUSZ and CHSZ datasets demonstrated that KDF-MutualSHOT outperformed other supervised and source-free domain adaptation approaches in cross-subject seizure subtype classification. © 2024, CC BY.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

FADE: A Task-Agnostic Upsampling Operator for Encoder-Decoder Architectures

arXiv

引用

arXiv 2024年

作者： Lu, Hao Liu, Wenze Fu, Hongtao Cao, Zhiguo The Key Laboratory of Image Processing and Intelligent Control Ministry of Education School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China

The goal of this work is to develop a task-agnostic feature upsampling operator for dense prediction where the operator is required to facilitate not only region-sensitive tasks like semantic segmentation but also detail-sensitive tasks such as image matting. Prior upsampling operators often can work well in either type of the tasks, but not both. We argue that task-agnostic upsampling should dynamically trade off between semantic preservation and detail delineation, instead of having a bias between the two properties. In this paper, we present FADE, a novel, plug-and-play, lightweight, and task-agnostic upsampling operator by fusing the assets of decoder and encoder features at three levels: i) considering both the encoder and decoder feature in upsampling kernel generation;ii) controlling the per-point contribution of the encoder/decoder feature in upsampling kernels with an efficient semi-shift convolutional operator;and iii) enabling the selective pass of encoder features with a decoder-dependent gating mechanism for compensating details. To improve the practicality of FADE, we additionally study parameter- and memory-efficient implementations of semi-shift convolution. We analyze the upsampling behavior of FADE on toy data and show through large-scale experiments that FADE is task-agnostic with consistent performance improvement on a number of dense prediction tasks with little extra cost. For the first time, we demonstrate robust feature upsampling on both region- and detail-sensitive tasks successfully. Code is made available at: https://***/poppinace/fade Copyright © 2024, The Authors. All rights reserved.

关键词： Object detection

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：