检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yang, Chen He, Yangfan Tian, Aaron Xuxiang Chen, Dong Wang, Jianhui Shi, Tianyu Heydarian, Arsalan Liu, Pei Department of Computer Science and Informatics Cardiff University United Kingdom College of Libera Arts University of Minnesota - Twin Cities United States United States Agricultural & Biological Engineering Mississippi State University United States Information and Software Engineering University of Electronic Science and Technology of China China Transportation Research Institute University of Toronto Canada Link Lab & Civil and Environmental Engineering University of Virginia United States Intelligent Transportation Thrust The Hong Kong University of Science and Technology Guangzhou China

In this paper, we introduce a novel approach for autonomous driving trajectory generation by harnessing the complementary strengths of diffusion probabilistic models (a.k.a., diffusion models) and transformers. Our proposed framework, termed the "World-centric Diffusion Transformer"(WcDT), optimizes the entire trajectory generation process, from feature extraction to model inference. To enhance the scene diversity and stochasticity, the historical trajectory data is first preprocessed into "Agent Move Statement" and encoded into latent space using Denoising Diffusion Probabilistic Models (DDPM) enhanced with Diffusion with Transformer (DiT) blocks. Then, the latent features, historical trajectories, HD map features, and historical traffic signal information are fused with various transformer-based encoders that is used to enhance the interaction of agents with other elements in the traffic scene. The encoded traffic scenes are then decoded by a trajectory decoder to generate multimodal future trajectories. Comprehensive experimental results show that the proposed approach exhibits superior performance in generating both realistic and diverse trajectories, showing its potential for integration into automatic driving simulation systems. Our code is available at https://***/yangchen1997/WcDT. © 2024, CC BY.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Interplay between Federated Learning and Explainable Artificial Intelligence: a Scoping Review

arXiv

引用

arXiv 2024年

作者： Lopez-Ramos, Luis M. Leiser, Florian Rastogi, Aditya Hicks, Steven Strümke, Inga Madai, Vince I. Budig, Tobias Sunyaev, Ali Hilbert, Adam Holistic Systems department Simula Metropolitan Center for Digital Engineering Oslo Norway Department of Economics and Management Karlsruhe Institute of Technology Germany Department of Neuroradiology University Hospital Bonn Germany Department of Computer Science Faculty of Informatics Norwegian University of Science and Technology Trondheim Norway QUEST Center for Responsible Research Charité - Universitätsmedizin Berlin Berlin Germany School of Computing and Digital Technology Birmingham City University Birmingham United Kingdom ETH Zurich Switzerland School of Computation Information and Technology Technical University of Munich Germany Charité - Universitätsmedizin Berlin Germany

The joint implementation of Federated learning (FL) and Explainable artificial intelligence (XAI) will allow training models from distributed data and explaining their inner workings while preserving important aspects of privacy. Towards establishing the benefits and tensions associated with their interplay, this scoping review maps those publications that jointly deal with FL and XAI, focusing on publications where an interplay between FL and model interpretability or post-hoc explanations was found. In total, 37 studies met our criteria, with more papers focusing on explanation methods (mainly feature relevance) than on interpretability (mainly algorithmic transparency). Most works used simulated horizontal FL setups involving 10 or fewer data centers. Only one study explicitly and quantitatively analyzed the influence of FL on model explanations, revealing a significant research gap. Aggregation of interpretability metrics across FL nodes created generalized global insights at the expense of node-specific patterns being diluted. 8 papers addressed the benefits of incorporating explanation methods as a component of the FL algorithm. Studies using established FL libraries or following reporting guidelines are a minority. More quantitative research and structured, transparent practices are needed to fully understand their mutual impact and under which conditions it happens. © 2024, CC BY.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Color-avoiding connected spanning subgraphs with minimum number of edges

arXiv

引用

arXiv 2023年

作者： Pintér, József Varga, Kitti Department of Stochastics Institute of Mathematics Budapest University of Technology and Economics Hungary HUN-REN-BME Stochastics Research Group Hungary Department of Computer Science and Information Theory Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics Hungary HUN-REN-ELTE Egerváry Research Group Hungary MTA-ELTE Momentum Matroid Optimization Research Group Hungary

We call a (not necessarily properly) edge-colored graph edge-color-avoiding connected if after the removal of edges of any single color, the graph remains connected. For vertex-colored graphs, similar definitions of color-avoiding connectivity can be given. In this article, we investigate the problem of determining the maximum number of edges that can be removed from either an edge- or a vertex-colored, color-avoiding connected graph so that it remains color- avoiding connected. First, we prove that this problem is NP-hard, and then, we give a polynomial-time approximation algorithm for it. To analyze the approximation factor of this algorithm, we determine the minimum number of edges of color-avoiding connected graphs on a given number of vertices and with a given number of colors. Furthermore, we also consider a generalization of edge-color-avoiding connectivity to matroids. © 2023, CC BY.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

FreeTumor: Large-Scale Generative Tumor Synthesis in Computed Tomography Images for Improving Tumor Recognition

arXiv

引用

arXiv 2025年

作者： Wu, Linshan Zhuang, Jiaxin Zhou, Yanning He, Sunan Ma, Jiabo Luo, Luyang Wang, Xi Ni, Xuefeng Zhong, Xiaoling Wu, Mingxiang Zhao, Yinghua Duan, Xiaohui Vardhanabhuti, Varut Rajpurkar, Pranav Chen, Hao Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong Tencent AI Lab Shenzhen China Department of Biomedical Informatics Harvard University Boston United States Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong Department of Radiology Shenzhen People’s Hospital Shenzhen China Department of Radiology The Third Affiliated Hospital of Southern Medical University Guangzhou China Department of Radiology Sun Yat-Sen Memorial Hospital Sun Yat-Sen University Guangzhou China Department of Diagnostic Radiology Li Ka Shing Faculty of Medicine The University of Hong Kong Hong Kong Department of Chemical and Biological Engineering The Hong Kong University of Science and Technology Hong Kong Division of Life Science The Hong Kong University of Science and Technology Hong Kong State Key Laboratory of Molecular Neuroscience The Hong Kong University of Science and Technology Hong Kong Shenzhen-Hong Kong Collaborative Innovation Research Institute The Hong Kong University of Science and Technology Shenzhen China

Tumor is a leading cause of death worldwide, with an estimated 10 million deaths attributed to tumor-related diseases every year. AI-driven tumor recognition unlocks new possibilities for more precise and intelligent tumor screening and diagnosis. However, the progress is heavily hampered by the scarcity of annotated datasets, which demands extensive annotation efforts by radiologists. To tackle this challenge, we introduce FreeTumor, an innovative Generative AI (GAI) framework to enable large-scale tumor synthesis for mitigating data scarcity. Specifically, FreeTumor effectively leverages a combination of limited labeled data and large-scale unlabeled data for tumor synthesis training. Unleashing the power of large-scale data, FreeTumor is capable of synthesizing a large number of realistic tumors on images for augmenting training datasets. To this end, we create the largest training dataset for tumor synthesis and recognition by curating 161,310 publicly available Computed Tomography (CT) volumes from 33 sources, with only 2.3% containing annotated tumors. To validate the fidelity of synthetic tumors, we engaged 13 board-certified radiologists in a Visual Turing Test to discern between synthetic and real tumors. Rigorous clinician evaluation validates the high quality of our synthetic tumors, as they achieved only 51.1% sensitivity and 60.8% accuracy in distinguishing our synthetic tumors from real ones. Through high-quality tumor synthesis, FreeTumor scales up the recognition training datasets by over 40 times, showcasing a notable superiority over state-of-the-art AI methods including various synthesis methods and foundation models. On average, FreeTumor improves the segmentation Dice scores by 6.7% and early tumor detection sensitivity by 16.4%. These findings indicate promising prospects of FreeTumor in clinical applications, potentially advancing tumor treatments and improving the survival rates of patients. Copyright © 2025, The Authors. All rights reserved.

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

Learning Neural Ordinary Equations for Forecasting Future Links on Temporal Knowledge Graphs

arXiv

引用

arXiv 2021年

作者： Han, Zhen Ding, Zifeng Ma, Yunpu Gu, Yujia Tresp, Volker Institute of Informatics LMU Munich Germany Corporate Technology Siemens AG Germany Department of Electrical and Computer Engineering Technical University of Munich Germany

There has been an increasing interest in inferring future links on temporal knowledge graphs (KG). While links on temporal KGs vary continuously over time, the existing approaches model the temporal KGs in discrete state spaces. To this end, we propose a novel continuum model by extending the idea of neural ordinary differential equations (ODEs) to multi-relational graph convolutional networks. The proposed model preserves the continuous nature of dynamic multi-relational graph data and encodes both temporal and structural information into continuous-time dynamic embeddings. In addition, a novel graph transition layer is applied to capture the transitions on the dynamic graph, i.e., edge formation and dissolution. We perform extensive experiments on five benchmark datasets for temporal KG reasoning, showing our model's superior performance on the future link forecasting task. Copyright © 2021, The Authors. All rights reserved.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

MANAGING UNEXPECTED FAILURES OF NUCLEAR POWER PLANTS BY INCORPORATING ORGANIZATIONAL FACTORS INTO PROBABILISTIC RISK ASSESSMENT

MANAGING UNEXPECTED FAILURES OF NUCLEAR POWER PLANTS BY INCO...

引用

2021 International Topical Meeting on Probabilistic Safety Assessment and Analysis, PSA 2021

作者： Roy, Andrea Farshadmanesh, Pegah Sakurahara, Tatsuya Kee, Ernie Mohaghegh, Zahra Research Laboratory Department of Nuclear Plasma and Radiological Engineering Grainger College of Engineering University of Illinois at Urbana-Champaign UrbanaIL61801 United States Department of Computer Science University of Illinois Urbana-Champaign United States Illinois Informatics Institute University of Illinois at Urbana-Champaign United States Beckman Institute for Advanced Science and Technology University of Illinois Urbana-Champaign United States

Organizational factors have been identified as key contributors to accidents and incidents in nuclear power plants (NPPs). The Data-Theoretic (DT) methodology, previously developed by some of the authors of this paper, helps quantify the impacts of organizational factors on plant risk scenarios in Probabilistic Risk Assessment (PRA). In the DT methodology, "data analytics" are guided by "theory," enhancing the completeness of "causality" derived from data and helping to avoid potentially misleading results from solely data-driven approaches. This paper advances the quantification of uncertainties (e.g., uncertainty associated with the PRA analyst's assessment of organizational factors) in the DT methodology. The DT methodology includes two submodules: (I) DT-BASE, which focuses on the development of detailed causal relationships based on a theory-building process and creates a baseline quantification utilizing analyst interpretation of information extracted from theoretical references and standards;(II) DT-SITE, which conducts data analytics (text mining) to update (add or delete links and nodes) the causal model structure developed in DT-BASE, quantify causal relationships using historical event databases, and update the results of baseline quantification in DT-BASE using Bayesian analysis. This paper adds uncertainty analysis to the DT-BASE submodule. The updated DT approach is implemented in a case study targeting NPP organizational "training quality" that influences the "training" performance shaping factor (PSF) in Human Reliability Analysis (HRA) associated with the plant PRA. © 2021 PSA. All Rights Reserved.

关键词： Nuclear power plants

来源：评论

学校读者我要写书评

暂无评论

Opinion Mining using Machine Learning Approaches: A Critical Study

Opinion Mining using Machine Learning Approaches: A Critical...

引用

2020 International Conference on Computing and Information technology, ICCIT 2020

作者： Ramakrishnan, Jayabrabu Mavaluru, DInesh Srinivasan, Karthik Mubarakali, Azath Narmatha, C. Malathi, G. Jazan University College of Computer Science and Information Technology Department of Information Technology and Security Saudi Arabia Saudi Electronic University College of Computing and Informatics Department of Information Technology Saudi Arabia King Khalid University College of Computer Science Department of CNE Saudi Arabia University of Tabuk Faculty of Computer and Information Technology Department of Computer Science Tabuk Saudi Arabia Vellore Institute of Technology School of Computing Science and Engineering Chennai India

ISBN: (纸本)9781728126807

With the advancement of internet technology, there is a large volume of information exists on the web for users of the internet. These users not just utilize the accessible resources on the web yet, in addition, give their feedback, consequently creating more useful data. Opinion mining (OM) is contemplated as a subfield of NLP, information retrieval, and data mining. It is the way toward extricating people's thoughts and observations from unformed content, which with respect to the development of online social media and the huge amount of user feedback, has gotten an effective, attractive and furthermore complex issue. The fundamental in OM is classifying the polarity of text as far as positive (good), neutral (surprise) or negative (bad). This paper is proposed to review the methods and applications of opinion mining. This paper presents a review of OM and related strategies and additionally discusses the applications and difficulties for OM with knowledge into previous researchers. © 2020 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Self-Sovereign Identity Empowered Non-Fungible Patient Tokenization for Patient-Centric Health Information Exchange Using Blockchain technology

SSRN

引用

SSRN 2022年

作者： Zhuang, Yan Shyu, Chi-Ren Hong, Shenda Li, Pengfei Zhang, Luxia National Institute of Health Data Science Peking University Beijing China Instititue for Data Science and Informatics Electrical Engineering and Computer Science Department University of Missouri ColumbiaMO United States Advanced Institute of Information Technology Peking University Hangzhou China

Background: Patient tokenization is a novel approach that allows anonymous patient-level linkage across healthcare facilities, minimizing the risk of breaching protected health information which is one of the major persistent challenges in health information exchange (HIE). Since most patient tokenization is developed and managed by third parties, centralized tokenization is fundamentally unable to address data security concerns. Non-Fungible Tokens (NFT), which are non-transferable cryptographic assets developed by leveraging blockchain’s unique properties, have the potential to provide secure, decentralized, and trustworthy patient tokenization. To verify the ownership of NFTs in a decentralized manner, Self-Sovereign Identity (SSI) is a user-centric approach that allows individuals full control over their *** procedures: We have developed a blockchain architecture that contains four modules: (1) Creation module, which permits users to create NFTs through blockchain, (2) Linkage module, which links the patient account from each health system to the patient NFT, (3) Authentication Module, which allows patients to permit healthcare providers to access their token to retrieve their patient’s permitted medical history from multiple healthcare facilities, and (4) Exchange Module, which involves validation of the legitimacy of the token through SSI and the HIE *** findings: A case study has been conducted on our proposed architecture from NFT creation to final HIE. Over 3 million transactions have been completed successfully with a blockchain validation and written time of 1.17 seconds on average. A stability test has also been conducted with a higher throughput of 200 transactions per second running for an hour with an average transaction processing time of 1.42 *** Conclusions: This study proposed a blockchain architecture that achieves SSI-enabled NFT-based patient tokenization. Our architecture design, implementation, and case stu

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Optimal Sizing of Security Constrained Unit Commitment Problem Integrated with Renewable Energy Sources and PEVs

Optimal Sizing of Security Constrained Unit Commitment Probl...

引用

International Conference on Electronics, computers and Artificial Intelligence (ECAI)

作者： Pravin G. Dhawale Vikram Kumar S.K. Bath Chaman Verma Maria Simona Raboaca Constantin Filote Deepak Kumar School of Electrical and Electronics Engineering Lovely Professional University Phagwara University of Calgary Calgary Canada Department of Electrical Engineering Maharaja Ranjit Singh Punjab Technical University Bhatinda India Department of Media and Educational Informatics Faculty of Informatics Eötvös Loránd University Budapest Hungary ICSI Energy Department National Research and Development Institute for Cryogenics and Isotopic Technologies Ramnicu Valcea Faculty of Electrical Engineering and Computer Science Stefan cel Mare University of Suceava Universitatii Street Suceava Romania Apex Institute of Technology Chandigarh Univerity Mohali Punjab India

Optimal sizing of SCUCP with solar energy and PEVs is crucial for electric power system operation and planning. This paper presents a Chaotic Arithmetic Optimization Algorithm to minimize cost while optimizing the sizing of solar energy and PEV systems. The proposed framework integrates probabilistic forecasting models for solar power generation and realistic EV charging profiles based on travel patterns and infrastructure characteristics. It aims to find optimal sizes for Solar power plants and EV charging infrastructure, considering system demand, cost, emissions, and reliability. The Chaotic Arithmetic Optimization Algorithm efficiently solves the optimization problem by accounting for the complexities of SCUCP. The research evaluates the proposed approach using benchmark test systems of varying dimensions and power systems. The results enable informed decision-making by identifying optimal capacities for solar power plants and charging infrastructure, leading to improved system reliability, reduced emissions, and minimized costs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Few-shot learning based histopathological image classification of colorectal cancer

引用

Intelligent Medicine 2024年第4期4卷 256-267页

作者： Rui Li Xiaoyan Li Hongzan Sun Jinzhu Yang Md Rahaman Marcin Grzegozek Tao Jiang Xinyu Huang Chen Li Key Laboratory of Intelligent Computing in Medical Image Ministry of EducationNortheastern UniversityShenyangLiaoning 110167China Cancer Hospital China Medical UniversityShenyangLiaoning 110122China Shengjing Hospital China Medical UniversityShenyang.Liaoning 110000China Institute for Medical Informatics University of LuebeckGermany Department of Knowledge Engineering University of Economics in KatowicePoland School of Intelligent Medicine Chengdu University of Traditional Chinese MedicineChengduSichuan 610075China International Joint Institute of Robotics and Intelligent Systems Chengdu University of Information TechnologyChengduSichuan 610225China School of Computer Science and Engineering University of New South WalesSydneyNSW 2052Australia

Background Colorectal cancer is a prevalent and deadly disease worldwide,posing significant diagnostic *** histopathologic image classification is often inefficient and *** some histopathologists use computer-aided diagnosis to improve efficiency,these methods depend heavily on exten-sive data and specific annotations,limiting their *** address these challenges,this paper proposes a method based on few-shot *** This study introduced a few-shot learning approach that combines transfer learning and contrastive learning to classify colorectal cancer histopathology images into benign and malignant *** model comprises modules for feature extraction,dimensionality reduction,and classification,trained using a combi-nation of contrast loss and cross-entropy *** this paper,we detailed the setup of hyperparameters:n-way,κ-shot,β,and the creation of support,query,and test *** Our method achieved over 98% accuracy on a query dataset with 35 samples per category using only 10 training samples per *** documented the model’s loss,accuracy,and the confusion matrix of the ***,we employed the t-SNE algorithm to analyze and assess the model’s classification *** The proposed model may demonstrate significant advantages in accuracy and minimal data depen-dency,performing robustly across all tested n-way,κ-shot *** consistently achieved over 93% accuracy on comprehensive test datasets,including 1916 samples,confirming its high classification accuracy and strong generalization *** research could advance the use of few-shot learning in medical diagnostics and also lays the groundwork for extending it to deal with rare,difficult-to-diagnose cases.

关键词： Colorectal cancer Few-shot learning Transfer learning Contrastive learning Histopathological images Benign and malignant categories

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：