检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2025年

作者： Jahin, Md Abrar Masud, Md. Akmol Suva, Md Wahiduzzaman Mridha, M.F. Dey, Nilanjan Okinawa Institute of Science and Technology Graduate University Okinawa904-0412 Japan Jahangirnagar University Institute of Information Technology Dhaka1342 Bangladesh American International University-Bangladesh Department of Computer Science Dhaka1229 Bangladesh Techno International New Town Department of Computer Science & Engineering New Town Kolkata700156 India

The rapid data surge from the high-luminosity Large Hadron Collider introduces critical computational challenges requiring novel approaches for efficient data processing in particle physics. Quantum machine learning, with its capability to leverage the extensive Hilbert space of quantum hardware, offers a promising solution. However, current quantum graph neural networks (GNNs) lack robustness to noise and are often constrained by fixed symmetry groups, limiting adaptability in complex particle interaction modeling. This paper demonstrates that replacing the classical Lorentz Group Equivariant Block modules in LorentzNet with a dressed quantum circuit significantly enhances performance despite using ≈5.5 times fewer parameters. Additionally, quantum circuits effectively replace MLPs by inherently preserving symmetries, with Lorentz symmetry integration ensuring robust handling of relativistic invariance. Our Lorentz-Equivariant Quantum Graph Neural Network (Lorentz-EQGNN) achieved 74.00% test accuracy and an AUC of 87.38% on the Quark-Gluon jet tagging dataset, outperforming the classical and quantum GNNs with a reduced architecture using only 4 qubits. On the Electron-Photon dataset, Lorentz-EQGNN reached 67.00% test accuracy and an AUC of 68.20%, demonstrating competitive results with just 800 training samples. Evaluation of our model on generic MNIST and FashionMNIST datasets confirmed Lorentz-EQGNN’s efficiency, achieving 88.10% and 74.80% test accuracy, respectively. Ablation studies validated the impact of quantum components on performance, with notable improvements in background rejection rates over classical counterparts. These results highlight Lorentz-EQGNN’s potential for immediate applications in noise-resilient jet tagging, event classification, and broader data-scarce HEP tasks. © 2020 IEEE.

关键词： Hilbert spaces

来源：评论

学校读者我要写书评

暂无评论

Structure-from-Sherds++: Robust Incremental 3D Reassembly of Axially Symmetric Pots from Unordered and Mixed Fragment Collections

arXiv

引用

arXiv 2025年

作者： Yoo, Seong Jong Liu, Sisung Arshad, Muhammad Zeeshan Kim, Jinhyeok Kim, Young Min Aloimonos, Yiannis Fermüller, Cornelia Joo, Kyungdon Kim, Jinwook Hong, Je Hyeong Department of Computer Science University of Maryland College ParkMD United States Department of Artificial Intelligence Hanyang University Korea Republic of Faculty of Health Sciences University of Ottawa ON Canada Artificial Intelligence Graduate School UNIST Korea Republic of Department of Electrical and Computer Engineering Seoul National University Korea Republic of Intelligence and Interaction Research Center Korea Institute of Science and Technology Korea Republic of Artificial Intelligence Graduate School The Department of Computer Science and Engineering UNIST Korea Republic of Department of Electronic Engineering Hanyang University Korea Republic of

Reassembling multiple axially symmetric pots from fragmentary sherds is crucial for cultural heritage preservation, yet it poses significant challenges due to thin and sharp fracture surfaces that generate numerous false positive matches and hinder large-scale puzzle solving. Existing global approaches, which optimize all potential fragment pairs simultaneously or data-driven models, are prone to local minima and face scalability issues when multiple pots are intermixed. Motivated by Structure-from-Motion (SfM) for 3D reconstruction from multiple images, we propose an efficient reassembly method for axially symmetric pots based on iterative registration of one sherd at a time, called Structure-from-Sherds++ (SfS++). Our method extends beyond simple replication of incremental SfM and leverages multi-graph beam search to explore multiple registration paths. This allows us to effectively filter out indistinguishable false matches and simultaneously reconstruct multiple pots without requiring prior information such as base or the number of mixed objects. Our approach achieves 87% reassembly accuracy on a dataset of 142 real fragments from 10 different pots, outperforming other methods in handling complex fracture patterns with mixed datasets and achieving state-of-the-art performance. Code and results can be found in our project page https://***/sfs/. Copyright © 2025, The Authors. All rights reserved.

关键词： Historic preservation

来源：评论

学校读者我要写书评

暂无评论

Brain Function Assessment of EEG in Mild Cognitive Impairment using Timelag Analysis

IEEJ Transactions on Electronics, Information and Systems

引用

IEEJ Transactions on Electronics, Information and Systems 2025年第4期145卷 467-473页

作者： Tanabe, Akifumi Harachi, Kento Yamamoto, Yusuke Muramatsu, Ayumi Nagahara, Hajime Takemura, Noriko Shimojo, Shinji Furihata, Daisuke Mizuno-Matsumoto, Yuko Graduate School of Information Science University of Hyogo 7-1-28 Minatojima Minami Chuo-ku Hyogo Kobe650-0047 Japan Department of Medical Engineering Aino University 4-5-4 Higashi-Ohda Ibaraki Osaka567-0012 Japan D3 Center The University of Osaka 2-8 Yamadaoka Osaka Suita565-0871 Japan D3 Center The University of Osaka 5-1 Mihogaoka Ibaraki Osaka567-0047 Japan Graduate School of Computer Science and Systems Engineering Kyushu Institute of Technology 680-4 Kawazu Fukuoka Iizuka820-8502 Japan Faculty of Software and Information Technology Aomori University 2-3-1 Shachihata Aomori-shi Aomori030-0943 Japan

This research aimed to reveal the information process in the brain of the patients who suffered from mild cognitive impairment (MCI) using the power spectra and timelag analysis in electroencephalogram (EEG). The patients with MCI have memory or thinking problems, and are at a great risk of developing Alzheimer's disease or a related dementia. Ten patients with MCI and 13 healthy adults (non-MCI) were examined. The EEGs (122.88 seconds/subject) during a state of relaxation with the eyes closed, were analyzed using the Fourier analysis (power spectrum) and the timelag analysis. The results showed that the mean value of the power spectra in MCI was significantly higher than the mean value in non-MCI in the theta band. The mean value of the absolute timelag values (ATL) in MCI was significantly lower than the mean value in non-MCI in the frontal and left temporal areas. Our findings suggest that the features from EEG of MCI were abstracted in the theta band, and the propagation of information related to memory in brain would becoming fast in MCI. © 2025 The Institute of Electrical Engineers of Japan.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Inter-radar Interference Mitigation via Self-attention Augmented biGRU and Customized Loss Functions

Inter-radar Interference Mitigation via Self-attention Augme...

引用

International Conference on Computing, Networking and Communications (ICNC)

作者： Yudai Suzuki Xiaoyan Wang Masahiro Umehira Hao Zhou Graduate School of Science and Engineering Ibaraki University Hitachi Japan Department of Electronics and Communication Technology Nanzan University Nagoya Japan School of Computer Science University of Science and Technology of China Hefei China

ISBN: (数字)9798331520960

ISBN: (纸本)9798331520977

CS (Chirp Sequence) radars are one of the most essential sensors for perceiving the surrounding environment in automotive applications. However, as the number of radar sensors increases, inter-radar interference becomes more prevalent, leading to a severe degradation in target detection rates. To address this issue, we propose a self-attention augmented dual-channel biGRU model to mitigate wideband interference, and explore various customized loss functions to enhance its performance. Specifically, the dual-channel biGRU processes the real and imaginary components of the radar beat signal simultaneously, while multi-head self-attention enhances the model’s focus on critical segments of the input signal. Additionally, the customized loss functions either incorporate both time and frequency domain information or utilize both interfered and interference-free samples to more effectively reconstruct the beat signal. The effectiveness of the proposed interference mitigation method is validated through extensive simulations, showing superior performance in terms of SNR (Signal-to-Noise Ratio), amplitude absolute error, phase absolute error, and F1 score compared to existing approaches.

关键词： Prevention and mitigation Radar detection Interference Radar Object detection Radar imaging Radar signal processing Sensors Wideband Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

ATP_mCNN: Predicting ATP binding sites through pretrained language models and multi-window neural networks

引用

computers in Biology and Medicine 2025年 185卷 109541-109541页

作者： Le, Van-The Malik, Muhammad-Shahid Lin, Yi-Jing Liu, Yu-Chen Chang, Yan-Yun Ou, Yu-Yen Department of Computer Science and Engineering Yuan Ze University Chung-Li32003 Taiwan Graduate Program in Biomedical Informatics Yuan Ze University Chung-Li32003 Taiwan Department of Computer Sciences Karakoram International University Gilgit-Baltistan 15100 Pakistan

Adenosine triphosphate plays a vital role in providing energy and enabling key cellular processes through interactions with binding proteins. The increasing amount of protein sequence data necessitates computational methods for identifying binding sites. However, experimental identification of adenosine triphosphate-binding residues remains challenging. To address the challenge, we developed a multi-window convolutional neural network architecture taking pre-trained protein language model embeddings as input features. In particular, multiple parallel convolutional layers scan for motifs localized to different window sizes. Max pooling extracts salient features concatenated across windows into a final multi-scale representation for residue-level classification. On benchmark datasets, our model achieves an area under the ROC curve of 0.95, significantly improving on prior sequence-based models and outperforming convolutional neural network baselines. This demonstrates the utility of pre-trained language models and multi-window convolutional neural networks for advanced sequence-based prediction of adenosine triphosphate-binding residues. Our approach provides a promising new direction for elucidating binding mechanisms and interactions from primary structure. © 2024 Elsevier Ltd

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Integration of data science with the intelligent IoT(IIoT): Current challenges and future perspectives

引用

Digital Communications and Networks 2025年第02期 280-298页

作者： Inam Ullah Deepak Adhikari Xin Su Francesco Palmieri Celimuge Wu Chang Choi Department of Computer Engineering Gachon University University of Electronic Science and Technology of China College of Information Science and Engineering Hohai University Department of Computer Science University of Salerno Graduate School of Informatics and Engineering The University of Electro-Communications

The Intelligent Internet of Things(IIoT) involves real-world things that communicate or interact with each other through networking technologies by collecting data from these “things” and using intelligent approaches, such as Artificial Intelligence(AI) and machine learning, to make accurate decisions. Data science is the science of dealing with data and its relationships through intelligent approaches. Most state-of-the-art research focuses independently on either data science or IIoT, rather than exploring their integration. Therefore, to address the gap, this article provides a comprehensive survey on the advances and integration of data science with the Intelligent IoT(IIoT) system by classifying the existing IoT-based data science techniques and presenting a summary of various characteristics. The paper analyzes the data science or big data security and privacy features, including network architecture, data protection, and continuous monitoring of data, which face challenges in various IoT-based systems. Extensive insights into IoT data security, privacy, and challenges are visualized in the context of data science for IoT. In addition, this study reveals the current opportunities to enhance data science and IoT market development. The current gap and challenges faced in the integration of data science and IoT are comprehensively presented, followed by the future outlook and possible solutions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Optimizing Multi-hop Distributed Collaborative Task Offloading in R2X

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Zhao, Wei Weng, Tangjie Cheng, Yu Liu, Zhi Kato, Nei Anhui University of Technology School of Computer Science and Technology Anhui Ma'anshan243032 China University of Electro-Communications Department of Computer and Network Engineering Tokyo Japan Tohoku University Graduate School of Information Sciences Sendai Japan

Roadside units (RSUs) with strong sensing abilities enhance the viability of the RSU-to-Everything (R2X) paradigm, offering crucial infrastructure support for Mobile Edge Computing (MEC) that enables real-time data processing and reduced delay. Since RSUs collect a large volume of data but have limited computing capability, data analysis tasks are usually offloaded to other network nodes, such as the cloud, other RSUs, or even vehicles. The multi-hop distributed collaborative task offloading scheme is expected to achieve high resource utilization efficiency and low task delay in this scenario, despite increasing energy consumption in data transmission. However, the highly dynamic nature of the R2X network topology makes it challenging for a node to independently select the next hop and collaboratively allocate tasks to neighbors in a multi-hop transmission path. Specifically, offloading decisions made by an individual node are influenced not only by its immediate neighbors but also by other nodes along the multi-hop path, referred to in this paper as the effect value. Additionally, the heterogeneity in computing resources and link delays among network nodes further increases the difficulty. To address these challenges, we first apply a Long Short-Term Memory (LSTM) model to predict and update the neighbors for each node while considering effect values, allowing them to independently adapt to environmental changes. Then, we design a two-layer Deep Reinforcement Learning (DRL) algorithm for network nodes to make decisions. The first-layer DRL algorithm is implemented by RSUs to determine task offloading modes. When an RSU decides to offload tasks to multiple vehicles for collaborative computing, the second-layer DRL algorithm is used by a vehicle to select its next hop vehicle and allocate tasks. Simulation results show that our proposed approach effectively adapts to topology changes in complex and highly dynamic network environments. Compared with existing methods,

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Learnable Non-uniform Quantization With Sampling-based Optimization for Variable-Rate Learned Image Compression

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Li, Shaohui Dai, Wenrui Kan, Nuowen Li, Chenglin Zou, Junni Xiong, Hongkai Tsinghua University Shenzhen International Graduate School Shenzhen Guangdong518055 China Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai200240 China Shanghai Jiao Tong University Department of Electronic Engineering Shanghai200240 China

Variable-rate coding is challenging but indispensable for learned image compression (LIC) that is in nature characterized by nonlinear transform coding (NTC). Existing methods for variable-rate LIC are restricted by the non-smooth quantization process with zero gradients almost everywhere, and consequently, suffer from training-test gap and degraded rate-distortion (R-D) performance. To address this problem, in this paper, we propose sampling-based optimization for training NTC models along with non-uniform quantizers. Different from gradient-based optimization, the proposed sampling-based optimization first randomly samples the parameters from Gaussian distributions with progressively reduced variance and then selects the optimal parameters with a R-D indicator. On the basis of sampling-based optimization, we develop a learnable non-uniform deadzone quantizer by adaptively refining the quantization steps for variable-rate coding with nonlinear transforms. Furthermore, we incorporate the learnable dead-zone quantizer to achieve a variable-rate LIC model with enhanced R-D performance and design rate and distortion control algorithms to adapt to dynamic network conditions. Experimental results show that the proposed method achieves state-of-the-art R-D performance in variable-rate image compression. It obtains an average 8.82% BD-rate reduction compared to latest versatile video compression (VVC) standard, and simultaneously achieves precise rate and distortion control with an average variation of 0.0087 bpp in bit-rates and 0.1265 dB in distortion on the Kodak dataset. © 2025 IEEE.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

Positioning Control for Voice Coil Motor Using Unbalanced Multilevel Inverter Topology

Positioning Control for Voice Coil Motor Using Unbalanced Mu...

引用

International Conference on Mechatronics (ICM)

作者： Gaku Takei Hidemine Obara Ryo Haneda Takahiro Nozaki Graduate School of Science and Technology Keio University Yokohama Japan Department of Physics Electrical and Computer Engineering Yokohama National University Yokohama Japan Department of System Design Engineering Keio University Yokohama Japan

ISBN: (数字)9798331533892

ISBN: (纸本)9798331533908

This article proposes a new multi-level inverter (MLI) based on flying capacitor inverter (FCI), with focused on high-precision positioning control for a voice coil motor (VCM). Multi-level inverters (MLIs) have the characteristic that increasing the number of output voltage levels reduces the output voltage error. However, as the number of levels increases, the number of circuit elements increases, and the circuit becomes more complex. In addition, the output voltage amplitude for each level is constant in conventional inverters. As a result, such as high-precision positioning control of VCM, designing an appropriate inverter for a reference wave that contains a large amount of low-voltage components increases the number of circuit elements. Therefore, this paper proposes a MLI with two unbalanced dc-input voltages, referred to as unbalanced MLIs (UMLIs) hereinafter. This topology uses two different dc-input voltages, and it is possible to reduce the output voltage level in extremely low-voltage range compared to other voltage levels. Simulation results when the high-precision positioning control of the VCM confirm that the proposed UMLIs improves the control performance compared to the conventional FCIs with equal number of output voltage levels and uniform level widths.

关键词： Coils Low voltage Simulation Switches Multilevel inverters Motors Topology Sensors Voltage control Motion control

来源：评论

学校读者我要写书评

暂无评论

Retrieval-Augmented Fine-Tuning With Preference Optimization For Visual Program Generation

arXiv

引用

arXiv 2025年

作者： Kang, Deokhyung Cho, Jeonghun Jeon, Yejin Jang, Sunbin Lee, Minsub Cho, Jawoon Lee, Gary Geunbae Graduate School of Artificial Intelligence POSTECH Korea Republic of Department of Computer Science and Engineering POSTECH Korea Republic of Hyundai Mobis T&I Company

Visual programming languages (VPLs) allow users to create programs through graphical interfaces, which results in easier accessibility and their widespread usage in various domains. To further enhance this accessibility, recent research has focused on generating VPL code from user instructions using large language models (LLMs). Specifically, by employing prompting-based methods, these studies have shown promising results. Nevertheless, such approaches can be less effective for industrial VPLs such as Ladder Diagram (LD). LD is a pivotal language used in industrial automation processes and involves extensive domain-specific configurations, which are difficult to capture in a single prompt. In this work, we demonstrate that training-based methods outperform prompting-based methods for LD generation accuracy, even with smaller backbone models. Building on these findings, we propose a two-stage training strategy to further enhance VPL generation. First, we employ retrieval-augmented fine-tuning to leverage the repetitive use of subroutines commonly seen in industrial VPLs. Second, we apply direct preference optimization (DPO) to further guide the model toward accurate outputs, using systematically generated preference pairs through graph editing operations. Extensive experiments on real-world LD data demonstrate that our approach improves program-level accuracy by over 10% compared to supervised fine-tuning, which highlights its potential to advance industrial automation. © 2025, CC BY.

关键词： Subroutines

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：