检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Hexiang Zhang Zhenghua Xu Dan Yao Shuo Zhang Junyang Chen Thomas Lukasiewicz State Key Laboratory of Reliability and Intelligence of Electrical Equipment School of Health Sciences and Biomedical Engineering Hebei University of Technology Tianjin China College of Computer Science and Software Engineering and Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) Shenzhen University Shenzhen China Institute of Logic and Computation TU Wien Vienna Austria Department of Computer Science University of Oxford Oxford United Kingdom

Analysis of X-ray images is one of the main tools to diagnose breast cancer. The ability to quickly and accurately detect the location of masses from the huge amount of image data is the key to reducing the morbidity and mortality of breast cancer. Currently, the main factor limiting the accuracy of breast mass detection is the unequal focus on the mass boxes, leading the network to focus too much on larger masses at the expense of smaller ones. In the paper, we propose the multi-head feature pyramid module (MHFPN) to solve the problem of unbalanced focus of target boxes during feature map fusion and design a multi-head breast mass detection network (MBMDnet). Experimental studies show that, comparing to the SOTA detection baselines, our method improves by 6.58% (in AP@50) and 5.4% (in TPR@50) on the commonly used IN-breast dataset, while about 6-8% improvements (in AP@20) are also observed on the public MIAS and BCS-DBT datasets.

关键词： Limiting Feature extraction Breast cancer Acoustics Speech processing X-ray imaging

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approach to Modeling and Exploring Random Sources of Gate-All-Around Silicon Nanosheet MOSFETs

Deep Learning Approach to Modeling and Exploring Random Sour...

引用

2022 International Symposium on VLSI Technology, systems and Applications, VLSI-TSA 2022

作者： Butola, Rajat Li, Yiming Kola, Sekhar Reddy National Yang-Ming Chiao Tung University Parallel and Scientific Computing Laboratory Hsinchu300 Taiwan National Yang-Ming Chiao Tung University Electrical Engineering and Computer Science International Graduate Program Hsinchu300 Taiwan National Yang-Ming Chiao Tung University Institute of Communications Engineering Hsinchu300 Taiwan National Yang-Ming Chiao Tung University Department of Electrical Engineering and Computer Engineering Hsinchu300 Taiwan National Yang-Ming Chiao Tung University Institute of Biomedical Engineering Hsinchu300 Taiwan Center for mmWave Smart Radar System and Technologies National Yang-Ming Chiao Tung University 1001 Ta-Hsueh Rd. Hsinchu300 Taiwan

ISBN: (纸本)9781665409230

In this paper, for the first time, deep learning (DL) based artificial neural network (ANN) is applied to model the effects of various random variations: work function fluctuation, random dopant fluctuation, and interface trap fluctuation, on gate-all-around silicon nanosheet MOSFETs. The number of fluctuations for each source variation is used as input features and their effects on devices of interest are studied qualitatively and quantitatively. The key figures of merit (FoM) are also extracted accurately from the transfer characteristics, which shows the competency of the ANN model in the domain of device modeling. © 2022 IEEE.

关键词： Nanosheets

来源：评论

学校读者我要写书评

暂无评论

GNN-Assisted BiG-AMP: Joint Channel Estimation and Data Detection for Massive MIMO Receiver

引用

IEEE Transactions on Wireless Communications 2025年第6期24卷 4631-4646页

作者： Liu, Zishen Wu, Nan He, Dongxuan Yuan, Weijie Li, Yonghui Quek, Tony Q. S. School of Information and Electronics Beijing Institute of Technology Beijing100081 China School of System Design and Intelligent Manufacturing Shenzhen Key Laboratory of Robotics and Computer Vision Southern University of Science and Technology Shenzhen518055 China School of Electrical and Information Engineering The University of Sydney SydneyNSW2008 Australia Singapore University of Technology and Design 487372 Singapore Department of Electronic Engineering Kyung Hee University Yongin17104 Korea Republic of

In this paper, we develop a graph neural network (GNN)-assisted bilinear inference approach to enhance the receiver performance of the MIMO system through message passing-based joint channel estimation and data detection (JCD). Specifically, based on the bilinear generalized approximate message passing (BiG-AMP) framework and conditional correlation of signal, we propose a GNN-assisted BiG-AMP (GNN-BiGAMP) approach, which integrates a GNN module into the data-detection-loop to compensate the inaccurate marginal likelihood approximation. By leveraging the coupling between the channel and received symbols, a bilinear GNN-assisted BiG-AMP (BiGNN-BiGAMP) JCD receiver is further proposed. This method incorporates two GNNs with similar graph representation into the bilinear posterior estimation loops, which not only compensates for approximation errors but also alleviates performance loss due to premature variance convergence, thereby enhancing the receiver performance significantly. To fully exploit the supervised information from channel estimation and data detection, we propose a multitask learning based training scheme, which coordinates GNNs with different tasks in two loops. Simulation results show that our proposed GNN-assisted JCD receivers significantly outperform other JCD counterparts in terms of both channel estimation and data detection. © 2025 IEEE. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Experimental Results of Federated Learning in Unmanned Aerial Vehicle Swarm

Experimental Results of Federated Learning in Unmanned Aeria...

引用

2021 IEEE International Conference on Space-Air-Ground Computing, SAGC 2021

作者： Shen, Yun Dong, Chao Qu, Yuben Ren, Weiqing Zhu, Xiaojun Ministry of Industry and Information Technology Nanjing University of Aeronautics and Astronautics The Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space China Shanghai Jiao Tong University Department of Computer Science and Engineering China

ISBN: (纸本)9781665400114

Unmanned aerial vehicles (UAVs) need to deploy machine learning (ML) models to execute complex tasks such as target tracking. For traditional centralized ML, UAVs need to transmit the raw data to the ground central server for model training. However, the transmission of raw data not only causes a large delay but also may incur sensitive information leakage. Federated learning (FL) as a distributed machine learning enables multiple UAVs to form a swarm for distributed model training. Nevertheless, most of current researches on FL of UAV swarm are based on theoretical architecture and simulations. In this paper, we focus on verifying the feasibility of deploying FL in a UAV swarm through practical experiments. Moreover, we further explore the impact of the distance change between the training node UAVs and the parameter server UAV on the training accuracy. The experimental results show that the UAV swarm can perform 15 rounds of global training in 5 minutes, with a training accuracy of 84.28%. © 2021 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Air-Ground Integrated Federated Learning: An Experimental Implementation

Air-Ground Integrated Federated Learning: An Experimental Im...

引用

2021 IEEE International Conference on Space-Air-Ground Computing, SAGC 2021

作者： Jing, Yuqian Dong, Chao Qu, Yuben Zhou, Fuhui Ministry of Industry and Information Technology Nanjing University of Aeronautics and Astronautics The Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space China Shanghai Jiao Tong University Department of Computer Science and Engineering China

ISBN: (纸本)9781665400114

With the envision of sixth generation (6G) networks technology, diverse artificial intelligence (AI) services are gradually developed from the network center to the edge, which enables unmanned aerial vehicle (UAV) as a hot spot to provide auxiliary services of machine learning (ML). Due to the sensitive privacy and limited resources, traditional centralized ML may not be used directly in such networks. As a promising distributed collaborative ML, federated learning (FL) is more suitable. However, the current research on air-ground integrated federal learning (AGIFL) mostly focuses on architecture design and network simulation. In this paper, we mainly verify the feasibility of deploying FL between UAV and terrestrial users through practical platform. Extensive experiments demonstrate that the training accuracy is higher when the training nodes is close to the server, and the accuracy is improved from 69.48% to 83.52% when distance is reduced from 20 meters to 5 meters. © 2021 IEEE.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Downstream-agnostic Adversarial Examples

arXiv

引用

arXiv 2023年

作者： Zhou, Ziqi Hu, Shengshan Zhao, Ruizhi Wang, Qian Zhang, Leo Yu Hou, Junhui Jin, Hai School of Cyber Science and Engineering Huazhong University of Science and Technology China School of Computer Science and Technology Huazhong University of Science and Technology China School of Cyber Science and Engineering Wuhan University China School of Information and Communication Technology Griffith University Australia Department of Computer Science City University of Hong Kong Hong Kong National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Hubei Key Laboratory of Distributed System Security China Hubei Engineering Research Center on Big Data Security China Cluster and Grid Computing Lab

Self-supervised learning usually uses a large amount of unlabeled data to pre-train an encoder which can be used as a general-purpose feature extractor, such that downstream users only need to perform fine-tuning operations to enjoy the benefit of "large model". Despite this promising prospect, the security of pre-trained encoder has not been thoroughly investigated yet, especially when the pre-trained encoder is publicly available for commercial use. In this paper, we propose AdvEncoder, the first framework for generating downstream-agnostic universal adversarial examples based on the pre-trained encoder. AdvEncoder aims to construct a universal adversarial perturbation or patch for a set of natural images that can fool all the downstream tasks inheriting the victim pre-trained encoder. Unlike traditional adversarial example works, the pre-trained encoder only outputs feature vectors rather than classification labels. Therefore, we first exploit the high frequency component information of the image to guide the generation of adversarial examples. Then we design a generative attack framework to construct adversarial perturbations/patches by learning the distribution of the attack surrogate dataset to improve their attack success rates and transferability. Our results show that an attacker can successfully attack downstream tasks without knowing either the pre-training dataset or the downstream dataset. We also tailor four defenses for pre-trained encoders, the results of which further prove the attack ability of AdvEncoder. Our codes are available at: https://github. com/CGCL-codes/AdvEncoder. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A Dimensional Structure based Knowledge Distillation Method for Cross-Modal Learning

arXiv

引用

arXiv 2023年

作者： Si, Lingyu Dong, Hongwei Qiang, Wenwen Yu, Junzhi Zhai, Wenlong Zheng, Changwen Xu, Fanjiang Sun, Fuchun Science & Technology on Integrated Information System Laboratory Institute of Software Chinese Academy of Sciences University of Chinese Academy of Sciences China College of Engineering Peking University China Department of Computer Science and Technology Tsinghua University China

Due to limitations in data quality, some essential visual tasks are difficult to perform independently. Introducing previously unavailable information to transfer informative dark knowledge has been a common way to solve such hard tasks. However, research on why transferred knowledge works has not been extensively explored. To address this issue, in this paper, we discover the correlation between feature discriminability and dimensional structure (DS) by analyzing and observing features extracted from simple and hard tasks. On this basis, we express DS using deep channel-wise correlation and intermediate spatial distribution, and propose a novel cross-modal knowledge distillation (CMKD) method for better supervised cross-modal learning (CML) performance. The proposed method enforces output features to be channel-wise independent and intermediate ones to be uniformly distributed, thereby learning semantically irrelevant features from the hard task to boost its accuracy. This is especially useful in specific applications where the performance gap between dual modalities is relatively large. Furthermore, we collect a real-world CML dataset to promote community development. The dataset contains more than 10, 000 paired optical and radar images and is continuously being updated. Experimental results on real-world and benchmark datasets validate the effectiveness of the proposed method. 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

A Review of Intelligent Vision Enhancement Technology for Battlefield

引用

Wireless Communications and Mobile Computing 2023年第1期2023卷

作者： Wu, Yufeng He, Yanjun Li, Ning Wang, Guowei Zhang, Longfei Zhang, Fuquan Zhu, Yifan The 15th Research Institute China Electronics Technology Group Corporation Beijing100083 China Key Laboratory of Digital Performance and Simulation Technology Beijing Institute of Technology Beijing100081 China College of Computer and Control Engineering Minjiang University Fuzhou350108 China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music Chengdu610021 China Department of Computer Science and Technology Tsinghua University Beijing100081 China

This paper reviews the application of intelligent vision enhancement technology in the battlefield environment and explores new research directions. This paper mainly introduces three parts. First, we introduce a solution for enhancing the battlefield situation and using head-mounted displays for soldier combat. Second, we summarize three core technologies supporting intelligent vision enhancement technology: 3D environment reconstruction, tracking and registration, and situational awareness. Third, we summarize three application directions of intelligent vision enhancement technology on the battlefield. Finally, the problems and challenges in the research work are proposed, including current issues such as accurate battlefield situational awareness by augmented reality technology, multiperson collaborative management and data flow, as well as challenges in future battlefield situation enhancement and perception. The future development trend of intelligent vision enhancement technology on the battlefield has been prospected. Two meanings of this paper are as follows: the first is to review the research status of intelligent vision enhancement technology from the technical level and identify the key technical points that may restrict development in the future;the second is to analyze the advantages and disadvantages of intelligent vision enhancement technology from the level of battlefield application and the roles of users. In addition, this paper proposes how to take the lead and take initiative in future wars. © 2023 Yufeng Wu et al.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Multi-kernel Correntropy Regression: Robustness, Optimality, and Application on Magnetometer Calibration

arXiv

引用

arXiv 2023年

作者： Li, Shilei Chen, Yihan Lou, Yunjiang Shi, Dawei Li, Lijing Shi, Ling The Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong The State Key Laboratory of Robotics and System School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen Shenzhen518055 China The School of Automation Beijing Institute of Technology China The School of Information and Control Engineering China University of Mining and Technology China

This paper investigates the robustness and optimality of the multi-kernel correntropy (MKC) on linear regression. We first derive an upper error bound for a scalar regression problem in the presence of arbitrarily large outliers and state that the kernel bandwidth plays an important role in minimizing the lowest upper error bound. Then, we find that the proposed MKC is related to a specific heavy-tail distribution, where its head shape is consistent with the Gaussian distribution but its tail shape is heavy-tailed, and the extent of heavy-tail is controlled by the kernel bandwidth. It becomes a Gaussian distribution when the bandwidth is infinite, which allows one to tackle both Gaussian and non-Gaussian problems without explicitly investigating the noise distributions. To explore the optimal underlying distribution parameters, an expectation-maximization-like (EM) algorithm is developed to estimate the parameter vectors and the distribution parameters in an alternating manner. The results show that our algorithm can achieve equivalent performance compared with the traditional linear regression under Gaussian noise, and it significantly outperforms the conventional method under heavy-tailed noise. Both numerical simulations and experiments on a magnetometer calibration application verify the effectiveness of the proposed method. Note to Practitioners-The goal of this paper is to enhance the accuracy of conventional linear regression in handling outliers while maintaining its optimality under Gaussian assumptions. Our algorithm is formulated under the maximum likelihood estimation (MLE) framework, assuming the regression residuals follow a type of heavy-tailed noise distribution with an extreme case of Gaussian. The degree of the heavy tail is explored alternatingly using an Expectation-Maximization (EM) algorithm which converges very quickly. The robustness and optimality of the proposed approach are investigated and compared with the traditional approaches. Both th

关键词： Linear regression

来源：评论

学校读者我要写书评

暂无评论

High Accuracy Source Localization Based on Parallel Factor Analysis of TDOA in the Cross Correlation Domain

引用

IEEE Transactions on Communications 2025年

作者： Li, Jianfeng Jiang, Hang Li, Yingying Zhou, Fuhui Wu, Qihui Quek, Tony Q. S. Al-Dhahir, Naofal Nanjing University of Aeronautics and Astronautics College of Electronic and Information Engineering Key Laboratory of Dynamic Cognitive System of Electromagnetic Spectrum Space Ministry of Industry and Information Technology Nanjing211106 China Nanjing University of Aeronautics and Astronautics College of Artificial Intelligence Nanjing China Singapore University of Technology and Design Singapore487372 Singapore Yonsei University Yonsei Frontier Laboratory Seoul03722 Korea Republic of University of Texas at Dallas Department of Electrical and Computer Engineering RichardsonTX75080 United States

It is challenging to ensure both high accuracy and low complexity when localizing radiation sources. To address this challenge, we propose two novel methods leveraging time difference of arrival (TDOA) measurements. Specifically, we introduce a TDOA estimation method and a direct position determination (DPD) method based on parallel factor (PARAFAC) analysis in the cross-correlation domain. Initially, multiple sensors synchronously capture the source signal, and the cross-correlation function between signals received from a reference sensor and other sensors is calculated. Then, the primary cross-spectrum data undergoes an expansion and integration process to establish the PARAFAC model. Through cross-spectrum expansion, virtual nodes are formed, which further improves the estimation performance. The TDOA estimates for each sensor are obtained by normalizing and extracting the phase from this matrix. Additionally, we introduce a novel DPD method tailored for multipath propagation scenarios. Simulations and real-world measurements demonstrate the superiority and effectiveness of our proposed methods compared with cutting-edge methods. © 1972-2012 IEEE.

关键词： Time difference of arrival

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：