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： Yiming Lei Zilong Li Junping Zhang Hongming Shan Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Institute of Science and Technology for Brain-Inspired Intelligence & MOE Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence & MOE Frontiers Center for Brain Science Fudan University

ISBN: (纸本)9798331314385

The explainability of deep neural networks (DNNs) is critical for trust and reliability in AI systems. Path-based attribution methods, such as integrated gradients (IG), aim to explain predictions by accumulating gradients along a path from a baseline to the target image. However, noise accumulated during this process can significantly distort the explanation. While existing methods primarily concentrate on finding alternative paths to circumvent noise, they overlook a critical issue: intermediate-step images frequently diverge from the distribution of training data, further intensifying the impact of noise. This work presents a novel Denoising Diffusion Path (DDPath) to tackle this challenge by harnessing the power of diffusion models for denoising. By exploiting the inherent ability of diffusion models to progressively remove noise from an image, DDPath constructs a piece-wise linear path. Each segment of this path ensures that samples drawn from a Gaussian distribution are centered around the target image. This approach facilitates a gradual reduction of noise along the path. We further demonstrate that DDPath adheres to essential axiomatic properties for attribution methods and can be seamlessly integrated with existing methods such as IG. Extensive experimental results demonstrate that DDPath can significantly reduce noise in the attributions—resulting in clearer explanations—and achieves better quantitative results than traditional path-based methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Acceleration method for target HRRP generation based on DDPM

Acceleration method for target HRRP generation based on DDPM

引用

Signal, information and Data processing (ICSIDP), IEEE International Conference on

作者： Xitai Shen Qiang Zhou Bingqian Yu Xin Zhang Yanhua Wang Liang Zhang Radar Technology Research Institute Beijing Institute of Technology Beijing China Zhengzhou Academy of Intelligent Technology Beijing Institute of Technology Zhengzhou China Beijing Key Laboratory of Embedded Real-Time Information: Processing Technology Beijing China Chongqing Innovation Center Beijing Institute of Technology Chongqing China

ISBN: (数字)9798331515669

ISBN: (纸本)9798331515676

The target high resolution range profile (HRRP) is the sum of the vector projected on the radar line of sight of the target scatters, which can reflect the radial structure characteristics of the target and is one of the important means of radar target recognition. Existing HRRP-based target recognition methods adopt the idea of machine learning classification, which usually require sufficient training samples to obtain satisfactory recognition performance. To solve the dilemma of insufficient radar HRRP data caused by factors such as non-cooperative target and complexity of data acquisition process, this paper uses denoising diffusion probabilistic models (DDPM) to generate HRRP data. Furthermore, to solve the problem that the DDPM generation speed is too slow to meet the actual needs, this paper designs a fast DDPM HRRP (FDH) data generation method based on progressive knowledge distillation. Finally, the experimental results show that the generation speed is increased by about 25 times compared with the initial model under the premise of losing 7.6% similarity.

关键词： Training Target recognition Noise reduction Radar Radar scattering Data collection Feature extraction Sampling methods Diffusion models Vectors

来源：评论

学校读者我要写书评

暂无评论

Multi-Terrain Radar HRRP Target Detection Using Contrastive Learning

Multi-Terrain Radar HRRP Target Detection Using Contrastive ...

引用

Signal, information and Data processing (ICSIDP), IEEE International Conference on

作者： Xinyang Wang Mingchen Yuan Jiaqi Liu Yang Li Yanhua Wang Liang Zhang Radar Technology Research Institute Beijing Institute of Technology Beijing China Chongqing Innovation Center Beijing Institute of Technology Chongqing China Zhengzhou Academy of Intelligent Technology Beijing Institute of Technology Zhengzhou China Beijing Key Laboratory of Embedded Real-Time Information: Processing Technology Beijing China

ISBN: (数字)9798331515669

ISBN: (纸本)9798331515676

As radar resolution improves, high-resolution radars have significantly enhanced target detection performance through the generation of high-resolution range profiles (HRRPs). However, when the range resolution is much smaller than the target, the target’s energy is dispersed across multiple range cells, leading to poor detection performance with traditional energy-based detection algorithms. Furthermore, existing detection algorithms are highly sensitive to clutter characteristics and exhibit poor adaptability to varying scenarios. To address these issues, this paper proposes a Constant False Alarm Rate (CFAR) HRRP detection algorithm based on contrastive learning for multiple terrains. The algorithm improves the contrastive learning loss function to achieve well-clustered representations of clutter samples during training. Test samples are evaluated by calculating the Mahalanobis distance between them and the training samples, and compared against detection thresholds to accomplish target detection. Testing results show that the proposed algorithm achieves superior detection performance across multiple terrains. Furthermore, to reach a 90% detection probability, the proposed method requires a signal-to-clutter ratio (SCR) reduction of 1 to 4 dB compared to the competing methods.

关键词： Training Energy resolution Radar detection Object detection Contrastive learning Radar Detection algorithms Clutter Signal to noise ratio Testing

来源：评论

学校读者我要写书评

暂无评论

Stress recognition based on graph structure representation of facial StO2

Stress recognition based on graph structure representation o...

引用

IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

作者： Jiaxiu Wang Xinyu Liu Yingjuan Jia Tong Chen Jingru Huang Zheyuan Zhang College of Electronic and Information Engineering Southwest University Chongqing China Institute of Legal Psychology and Intelligent Computing Southwest University Chongqing China Chongqing Key Laboratory of Non-Linear Circuits and Intelligent Information Processing Southwest University Chongqing China

Stress is an integral state that affects physical and mental health of individuals. Tissue Oxygen Saturation (StO2) is an emerging physiological signal and can reflect different stress states. Previous studies have simply extracted features of StO2 from specific regions of the face without exploring potential associations between them. In this paper, we analyze the facial Region of Interest (ROI) based on StO2, so as to explore the deep relationship between ROIs. Building upon this, we construct, for the first time, a StO2-based facial graph structure for stress recognition by using ROIs as nodes. In addition, we propose a new graph pooling method called FTPool which allows to measure node significance in terms of both node characteristics and graph topology. Finally, we further propose a dual-stream network (GCNet) combining GNN and CNN for the baseline-independent stress recognition. The experimental results show that the proposed GCNet can achieve state-of-the-art (SOTA) results with an accuracy of 78.57% on the original unbalanced database.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Ship detection based on the modified YOLOV5 for SAR imagery

Ship detection based on the modified YOLOV5 for SAR imagery

引用

2024 International Conference on Image, Signal processing, and Pattern Recognition, ISPP 2024

作者： Wang, Zhongbo He, Miao Ding, Qinghai Luo, Haibo Key Laboratory of Opto-Electronic Information Processing Chinese Academy of Sciences Shenyang110016 China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China Space Star Technology Co Ltd. Beijing100086 China

ISBN: (纸本)9781510680425

Synthetic aperture radar (SAR) is an essential tool for ocean *** the main participants on the ocean, ships are the most important targets for ocean *** it is of great importance to develop ship detection algorithms for SAR sea *** algorithms based on Convolutional Neural Network perform far better than the traditional methods based on manual features on ship detection task due to the powerful feature representation *** algorithms based on Convolutional Neural Network can be divided into one-stage algorithms, and two-stage ***-stage algorithms have high accuracy, but are relatively ***-stage algorithms have high inference speed, but compared with two-stage algorithms, they have lower *** in this article, we proposed an modified one-stage detection algorithm to improve the accuracy of ship detection in a condition that the modified algorithm meet the real-time ***, the small model of one-stage algorithm YOLOV5 is chosen as the base network to get the high inference ***, to improve the accuracy of ship detection with a little increase in inference time and the model parameters, we integrate the one-layer super-resolution architecture with the simplist structure into the YOLOV5 ***, we conducted the comparative experiments on our Dataset to verify the performance of modified YOLO *** experimental results show that the modified method has obtained an Average Precision (AP) improvement than the original YOLO V5 for detecting ships in SAR images with a little increase in inference time and the model parameters. © 2024 SPIE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Graph Convolutional Network Robustness Verification Algorithm Based on Dual Approximation 1

引用

25th International Conference on Formal Engineering Methods, ICFEM 2024

作者： An, Dongdong Zhang, Hao Zhao, Qin Liu, Jing Shi, Jianqi Huang, Yanhong Yang, Yang Liu, Xu Qin, Shengchao Shanghai Engineering Research Center of Intelligent Education and Bigdata Shanghai Normal University Shanghai200234 China Shanghai Key Laboratory of Trustworthy Computing East China Normal University Shanghai200062 China Shanghai Formal-Tech Information Technology Co. LTD. Shanghai China Guangzhou Institute of Technolog Xidian University Shaanxi China ICCT and ISN Laboratory Xidian University Shaanxi China

ISBN: (数字)9789819606177

ISBN: (纸本)9789819606160

With the continuous development of Graph Neural Network (GNN) technologies, securing their robustness is crucial for their broad adoption in practical applications. Although various verification methods for training GNNs have been proposed, studies indicate that Graph Convolutional Networks (GCNs) remain vulnerable to adversarial attacks affecting both graph structure and node attributes. We propose a novel approach to verify the robustness of GCNs against perturbations in node attributes by employing a dual approximation technique to convexify nonlinear activation functions. This transformation changes the original non-convex problem into a more manageable convex forms. We start by applying linear relaxation to convert fixed-value features in each GCN layer into variables suitable for optimization. Next, we reframe the task of identifying the worst-case margin for a graph as a linear problem, which we solve using linear programming techniques. Given the discrete nature of graph data, we define a perturbation space that extends the data domain from discrete to continuous values. To improve the accuracy of the convex relaxation, we use a dual approximation algorithm to set bounds on the optimizable variables. Our method certifies the robustness of nodes against perturbations within a specified range and significantly improves verification accuracy compared to previous approaches. This method surpasses previous work in verification accuracy and is distinctively tailored to address the S-curve, an aspect less explored in prior research. Experimental results show that our method significantly refines the precision of robustness verification for GCNs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Operating Health State Prediction and Evaluation of Excitation Unit Based on GMM and LSTM 3rd

Operating Health State Prediction and Evaluation of Excitati...

引用

3rd China intelligent Robotics Annual Conference, CCF CIRAC 2022

作者： Ma, Yinxing Yang, Peihao Lv, Gang Deng, Shibin Jiao, Shangbin Li, Yujun Wu, Xiaohui Zhang, Jing Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi’an University of Technology Xi’an710048 China Xi’an Thermal Power Research Institute Co. Ltd. Xi’an710054 China Huaneng Weihai Power Generation Co. Ltd. Wei’hai264200 China

ISBN: (纸本)9789819903009

The excitation unit is the weak link in the generator operation since it produces internal deterioration under long-term operation. To evaluate the health condition of the key excitation unit and generate warning information in advance, this work proposes a health assessment method based on the Gaussian mixture model (GMM) and LSTM neural network. Firstly, the weights of each type of data will be determined by obtaining the electrical parameter data of the generator excitation unit from the SIS system, using the infrared image data of carbon brush and excitation transformer based on infrared image acquisition system, and the correlation between each electrical parameter data and temperature on account of the MIC correlation analysis method. On the top of it, LSTM neural network prediction model is established. And establish a GMM-based health benchmark model based on the data collected in the field during healthy operation, and design the health decline index HDI using the Marxist distance for evaluating the health status and deterioration degree of the unit. Finally, the method of this paper is validated by the 2022 operation data of a thermal power plant in Weihai, which proved that this method can provide a basis for regular maintenance and early identification of faults in the power plan and quantifying the degradation degree of the excitation unit under long-term operation and making early warning before the abnormal occurrence of the unit. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Deterioration

来源：评论

学校读者我要写书评

暂无评论

Incremental attribute reduction based on knowledge granularity under incomplete data 3

Incremental attribute reduction based on knowledge granulari...

引用

2021 3rd International Conference on Artificial Intelligence and Computer Science, AICS 2021

作者： Zhong, Qiangqiang Wang, Lei Yang, Wen Liu, Chao School of Information Engineering Nanchang Institute of Technology Nanchang330000 China Jiangxi Provincial Key Laboratory of Water Information Cooperative Sensing and Intelligent Processing Nanchang Institute of Technology Nanchang330000 China

Traditional attribute reductions can’t be directly applied to miss attribute values and dynamic changes data. Therefore, how to dynamically update attribute reductions efficiently when incomplete data changes dynamically is great research value. For the change of the object, using matrix’s simple and clear characteristics, the paper focus on the incremental attribute reduction algorithm in matrix form of knowledge granularity. Firstly, the knowledge granularity is obtained by matrix which uses to measure the uncertainty of the system, then calculating the attribute importance, and using the attribute importance to devise a static algorithm, and then analyzing incremental matrix form when the object increases. The comparative experiments show that the incremental algorithm can effectively and efficiently update the result of attribute reduction as the object changes. © Journal of Physics: Conference Series 2021.

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

Neural 3D Scene Reconstruction from Multiple 2D Images without 3D Supervision

arXiv

引用

arXiv 2023年

作者： Guo, Yi Sun, Che Jia, Yunde Wu, Yuwei Beijing Key Laboratory of Intellegent 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

Neural 3D scene reconstruction methods have achieved impressive performance when reconstructing complex geometry and low-textured regions in indoor scenes. However, these methods heavily rely on 3D data which is costly and time-consuming to obtain in real world. In this paper, we propose a novel neural reconstruction method that reconstructs scenes using sparse depth under the plane constraints without 3D supervision. We introduce a signed distance function field, a color field, and a probability field to represent a scene. We optimize these fields to reconstruct the scene by using differentiable ray marching with accessible 2D images as supervision. We improve the reconstruction quality of complex geometry scene regions with sparse depth obtained by using the geometric constraints. The geometric constraints project 3D points on the surface to similar-looking regions with similar features in different 2D images. We impose the plane constraints to make large planes parallel or vertical to the indoor floor. Both two constraints help reconstruct accurate and smooth geometry structures of the scene. Without 3D supervision, our method achieves competitive performance compared with existing methods that use 3D supervision on the ScanNet dataset. Copyright © 2023, The Authors. All rights reserved.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

Attention-based UAV Trajectory Optimization for Wireless Power Transfer-assisted IoT Systems

arXiv

引用

arXiv 2025年

作者： Dong, Li Jiang, Feibo Peng, Yubo School of Computer Science Hunan University of Technology and Business Changsha410205 China Xiangjiang Laboratory Changsha410205 China Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing Hunan Normal University Changsha410081 China School of Intelligent Software and Engineering Nanjing University Suzhou215163 China

Unmanned Aerial Vehicles (UAVs) in Wireless Power Transfer (WPT)-assisted Internet of Things (IoT) systems face the following challenges: limited resources and suboptimal trajectory planning. Reinforcement learning-based trajectory planning schemes face issues of low search efficiency and learning instability when optimizing large-scale systems. To address these issues, we present an Attention-based UAV Trajectory Optimization (AUTO) framework based on the graph transformer, which consists of an Attention Trajectory Optimization Model (ATOM) and a Trajectory lEarNing Method based on Actor-critic (TENMA). In ATOM, a graph encoder is used to calculate the self-attention characteristics of all IoTDs, and a trajectory decoder is developed to optimize the number and trajectories of UAVs. TENMA then trains the ATOM using an improved Actor-Critic method, in which the real reward of the system is applied as the baseline to reduce variances in the critic network. This method is suitable for high-quality and large-scale multi-UAV trajectory planning. Finally, we develop numerous experiments, including a hardware experiment in the field case, to verify the feasibility and efficiency of the AUTO framework. Copyright © 2025, The Authors. All rights reserved.

关键词： Antennas

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：