检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Cheng, Sibo Bocquet, Marc Ding, Weiping Finn, Tobias Sebastian Fu, Rui Fu, Jinlong Guo, Yike Johnson, Eleda Li, Siyi Liu, Che Moro, Eric Newton Pan, Jie Piggott, Matthew Quilodran, Cesar Sharma, Prakhar Wang, Kun Xiao, Dunhui Xue, Xiao Zeng, Yong Zhang, Mingrui Zhou, Hao Zhu, Kewei Arcucci, Rossella CEREA ENPC EDF R&D Institut Polytechnique de Paris Île-de-France France School of Artificial Intelligence and Computer Science Nantong University Jiangsu Nantong226019 China School of Mathematical Sciences Key Laboratory of Intelligent Computing and Applications Tongji University Shanghai200092 China School of Engineering and Materials Science Faculty of Science and Engineering Queen Mary University of London LondonE1 4NS United Kingdom Zienkiewicz Centre for Modelling Data and AI Faculty of Science and Engineering Swansea University SwanseaSA1 8EN United Kingdom Department of Computer Science and Engineering Hong Kong university of science and technology Hong Kong Department of Earth Science & Engineering Imperial College London LondonSW7 2AZ United Kingdom Tianjin Key Laboratory of Imaging and Sensing Microelectronics Technology School of Microelectronics Tianjin University Tianjin300072 China Centre for Health Informatics Cumming School of Medicine University of Calgary CalgaryABT2N 1N4 Canada Department of Community Health Sciences Cumming School of Medicine University of Calgary CalgaryABT2N 1N4 Canada Undaunted Grantham Institute for Climate Change and the Environment Imperial College London LondonSW7 2AZ United Kingdom Culham Campus AbingdonOX14 3DB United Kingdom Centre for Computational Science Department of Chemistry University College London LondonWC1E 6BT United Kingdom Concordia Institute for Information Systems Engineering Concordia University MontrealQCH3G 1M8 Canada School of Mechanical Medical and Process Engineering Faculty of Engineering Queensland University of Technology BrisbaneQLD Australia Department of Chemical Engineering University College London LondonWC1E 6BT United Kingdom

Unstructured grid data are essential for modelling complex geometries and dynamics in computational physics. Yet, their inherent irregularity presents significant challenges for conventional machine learning (ML) techniques. This paper provides a comprehensive review of advanced ML methodologies designed to handle unstructured grid data in high-dimensional dynamical systems. Key approaches discussed include graph neural networks, transformer models with spatial attention mechanisms, interpolation-integrated ML methods, and meshless techniques such as physics-informed neural networks. These methodologies have proven effective across diverse fields, including fluid dynamics and environmental simulations. This review is intended as a guidebook for computational scientists seeking to apply ML approaches to unstructured grid data in their domains, as well as for ML researchers looking to address challenges in computational physics. It places special focus on how ML methods can overcome the inherent limitations of traditional numerical techniques and, conversely, how insights from computational physics can inform ML development. To support benchmarking, this review also provides a summary of open-access datasets of unstructured grid data in computational physics. Finally, emerging directions such as generative models with unstructured data, reinforcement learning for mesh generation, and hybrid physics-data-driven paradigms are discussed to inspire future advancements in this evolving field. © 2025, CC BY-NC-SA.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Improving Knowledge Distillation via Category Structure 1

引用

16th European Conference on computer Vision, ECCV 2020

作者： Chen, Zailiang Zheng, Xianxian Shen, Hailan Zeng, Ziyang Zhou, Yukun Zhao, Rongchang School of Computer Science and Engineering Central South University ChangshaHunan410083 China Centre for Medical Image Computing University College London LondonWC1V 6LJ United Kingdom

ISBN: (数字)9783030586041

ISBN: (纸本)9783030586034

Most previous knowledge distillation frameworks train the student to mimic the teacher’s output of each sample or transfer cross-sample relations from the teacher to the student. Nevertheless, they neglect the structured relations at a category level. In this paper, a novel Category Structure is proposed to transfer category-level structured relations for knowledge distillation. It models two structured relations, including intra-category structure and inter-category structure, which are intrinsic natures in relations between samples. Intra-category structure penalizes the structured relations in samples from the same category and inter-category structure focuses on cross-category relations at a category level. Transferring category structure from the teacher to the student supplements category-level structured relations for training a better student. Extensive experiments show that our method groups samples from the same category tighter in the embedding space and the superiority of our method in comparison with closely related works are validated in different datasets and models. © 2020, Springer Nature Switzerland AG.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Real-time frame rate controlled dynamic voltage and frequency scaling for GPU power consumption optimization

引用

AIP Conference Proceedings 2024年第1期3131卷

作者： Guruprasad Konnurmath Satyadhyan Chickerur School of Computer Science and Engineering K L E Technological University Hubballi 580031 India Centre for High Performance Computing K L E Technological University Hubballi 580031 India

Technological advancements in the domain of computer graphics have enhanced the visual realism in video games, movies, and various forms of media. This comprises of realistic visual presentation, best rendering mechanisms, improved textures, naturalistic light effects etc. This paper focus on frame rate monitoring integrated with dynamic voltage and frequency scaling (DVFS) technique to reduce power consumption on rendering pipeline mainly the vertex and pixel shaders during 3D scene render process. To understand shader power analysis, power consumption on GPU is investigated initially at runtime to set rendering configuration by limiting framerates of rendered scene and generates dynamic power-aware rendering framework by analyzing the workload on frames. In the implementation, optimized power on GPU is attained with the combination of three major mechanisms: the TensorFlow tools and libraries, the Graphics Rendering Parameter Control mechanism and the Dynamic Voltage and Frequency Control mechanism. Initially the range of framerates are set in-between 30 fps and 60 fps to consider the state as graphics rendering parameter control mechanism. Based on workload of each frame, if current frame rate is greater than range of 60 fps then GPU frequency is reduced and if lesser than 30 fps then frequency is increased using DVFS mechanism. So that optimal power settings are achieved, avoiding unnecessary power wastage on GPU. parallelly TensorFlow libraries boost computational speed improving time and memory frequencies. Results show efficient render of 3D scene with power savings around 35% and decreased computational complexity. The results obtained through the proposed methodology forms the basis for performance and power consumption analysis of various other stages of graphics rendering pipeline. Minimal power consumption per frame-rate is achieved resulting into the best performance-power oriented results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Systematic Analysis of Meteorological Parameters in Predicting Rainfall Events

引用

IEEE Access 2025年

作者： Pathan, Muhammad Salman Nadella, Pardhu Haq, Yasin Ul Chaudhury, Soumini Wu, Jiantao Nag, Avishek Dev, Soumyabrata Dublin City University School of Computing Ireland Gokaraju Rangaraju Institute of Engineering and Technology Hyderabad India University of Engineering and Technology Lahore Narowal Campus Department of Computer Science and Engineering Punjab Narowal Pakistan Mrinalini Datta Mahavidyapith Kolkata India ADAPT SFI Research Centre Dublin Ireland University College School of Computer Science Dublin Ireland

Accurate rainfall prediction is of paramount significance across diverse sectors, particularly in agriculture, where accurate predictions play a pivotal role in effective resource management and decision-making. However, due to the complexity and dynamic structure of climate systems, rainfall prediction is a difficult task. This study shifts the focus towards exploring correlations and feature selection in the context of rainfall prediction, contributing to a more sophisticated understanding of the process. By analyzing five years of weather data from three weather stations in the United States, Canada, and Ireland, the study delves into the interactions between meteorological features and rainfall occurrences. The use of a machine learning (ML)-based feature importance technique, which enables the identification of key meteorological features that significantly contribute to rainfall prediction, is central to the work. As a result, this method improves understanding of meteorological conditions, which act as accurate forecasters of rainfall outcomes and can help to develop accurate decision-support systems. The study also conducts a thorough assessment of prediction performance of various ML and deep learning (DL) techniques such as Classification and Regression Trees (CART), Support Vector Machine (SVM) and Dense Neural Networks (DNN).The findings show that the models using only the important meteorological features in the dataset perform better than using all the features. This rigorous examination also supports the selection of appropriate rainfall forecast models for specific use cases. Overall, this study increases our understanding of rainfall prediction by focusing on the investigation of correlations between meteorological indicators and the identification of key meteorological features using ML approaches, offering valuable insights for weather forecasting applications. This nuanced analysis contributes to the advancement of predictive modeling in the real

关键词： Rain

来源：评论

学校读者我要写书评

暂无评论

Unravelling quantum chaos using persistent homology

arXiv

引用

arXiv 2022年

作者： Cao, Harvey Leykam, Daniel Angelakis, Dimitris G. Centre for Quantum Technologies National University of Singapore 3 Science Drive 2 Singapore117543 Singapore School of Electrical and Computer Engineering Technical University of Crete Chania73100 Greece AngelQ Quantum Computing 531A Upper Cross Street #04-95 Hong Lim Complex 051531 Singapore

Topological data analysis is a powerful framework for extracting useful topological information from complex datasets. Recent work has shown its application for the dynamical analysis of classical dissipative systems through a topology-preserving embedding method that allows reconstructing dynamical attractors, the topologies of which can be used to identify chaotic behaviour. Open quantum systems can similarly exhibit non-trivial dynamics, but the existing toolkit for classification and quantification are still limited, particularly for experimental applications. In this work, we present a topological pipeline for characterizing quantum dynamics, which draws inspiration from the classical approach by using single quantum trajectory unravelings of the master equation to construct analogue ‘quantum attractors’ and extracting their topology using persistent homology. We apply the method to a periodically modulated Kerr-nonlinear cavity to discriminate parameter regimes of regular and chaotic phase using limited measurements of the system. Copyright © 2022, The Authors. All rights reserved.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Correction: AI content detection in the emerging information ecosystem: new obligations for media and tech companies

引用

Ethics and Information Technology 2024年第4期26卷 1-2页

作者： Knott, Alistair Pedreschi, Dino Jitsuzumi, Toshiya Leavy, Susan Eyers, David Chakraborti, Tapabrata Trotman, Andrew Sundareswaran, Sundar Baeza-Yates, Ricardo Biecek, Przemyslaw Weller, Adrian Teal, Paul D. Basu, Subhadip Haklidir, Mehmet Morini, Virginia Russell, Stuart Bengio, Yoshua Social Media Governance Project Global Partnership on AI Montreal Canada School of Engineering and Computer Science Victoria University of Wellington Wellington New Zealand University of Pisa Pisa Italy Chuo University Tokyo Japan Insight SFI Research Centre for Data Analytics School of Information and Communication University College Dublin Dublin Ireland School of Computing University of Otago Dunedin New Zealand Alan Turing Institute London United Kingdom University College London London United Kingdom Institute for Experiential AI Northeastern University Silicon Valley USA Warsaw University of Technology Warsaw Poland University of Cambridge Cambridge United Kingdom Computer Science and Engineering Department Jadavpur University Kolkata India Artificial Intelligence Institute Tubitak Bilgem Gebze Türkiye Center for Human-Compatible AI UC Berkeley Berkeley USA Mila - Quebec AI Institute Montreal Canada University of Montreal Montreal Canada

来源：评论

学校读者我要写书评

暂无评论

Assessing responsible innovation training

引用

Journal of Responsible Technology 2023年 16卷

作者： Stahl, Bernd Carsten Aicardi, Christine Brooks, Laurence Craigon, Peter J. Cunden, Mayen Burton, Saheli Datta Heaver, Martin De Saille, Stevienna De Dolby, Serena Dowthwaite, Liz Eke, Damian Hughes, Stephen Keene, Paul Kuh, Vivienne Portillo, Virginia Shanley, Danielle Smallman, Melanie Smith, Michael Stilgoe, Jack Ulnicane, Inga Wagner, Christian Webb, Helena School of Computer Science University of Nottingham United Kingdom Dept. of Informatics and School of Biomedical Engineering & Imaging Sciences King's College London United Kingdom Information School University of Sheffield Sheffield United Kingdom Horizon Digital Economy Research School of Computer Science University of Nottingham United Kingdom Independent Researcher Dept of Science and Technology Studies University College London United Kingdom ORBIT RRI limited Leicester United Kingdom Dept of Sociological Studies University of Sheffield Sheffield United Kingdom Centre for Computing and Social Responsibility De Montfort University Leicester United Kingdom School of Chemistry University of Bristol United Kingdom Department of Philosophy Faculty of Arts and Social Sciences Maastricht University Netherlands Biodiversity and Conservation Science Department of Biodiversity Conservation and Attractions Wildlife Research Centre Western Australia De Montfort University United Kingdom Lab for Uncertainty in Data and Decision Making (LUCID) School of Computer Science University of Nottingham United Kingdom

There is broad agreement that one important aspect of responsible innovation (RI) is to provide training on its principles and practices to current and future researchers and innovators, notably including doctoral students. Much less agreement can be observed concerning the question of what this training should consist of, how it should be delivered and how it could be assessed. The increasing institutional embedding of RI leads to calls for the alignment of RI training with training in other subjects. One can therefore observe a push towards the official assessment of RI training, for example in the recent call for proposals for centres for doctoral training by UK Research and Innovation. This editorial article takes its point of departure from the recognition that the RI community will need to react to the call for assessment of RI training. It provides an overview of the background and open questions around RI training and assessment as a background of examples of RI training assessment at doctoral level. There is unlikely to be one right way of assessing RI training across institutions and disciplines, but we expect that the examples provided in this article can help RI scholars and practitioners orient their training and its assessment in ways that are academically viable as well as supportive of the overall aims of RI. © 2023

关键词： Assessment Doctoral training Education Responsible innovation

来源：评论

学校读者我要写书评

暂无评论

Short-Term and Long-Term Context Aggregation Network for Video Inpainting 16th

Short-Term and Long-Term Context Aggregation Network for Vid...

引用

16th European Conference on computer Vision, ECCV 2020

作者： Li, Ang Zhao, Shanshan Ma, Xingjun Gong, Mingming Qi, Jianzhong Zhang, Rui Tao, Dacheng Kotagiri, Ramamohanarao School of Computing and Information Systems The University of Melbourne Melbourne Australia UBTECH Sydney AI Centre School of Computer Science Faculty of Engineering The University of Sydney Darlington2008 Australia School of Information Technology Deakin University Geelong Australia School of Mathematics and Statistics The University of Melbourne Melbourne Australia

ISBN: (纸本)9783030585471

Video inpainting aims to restore missing regions of a video and has many applications such as video editing and object removal. However, existing methods either suffer from inaccurate short-term context aggregation or rarely explore long-term frame information. In this work, we present a novel context aggregation network to effectively exploit both short-term and long-term frame information for video inpainting. In the encoding stage, we propose boundary-aware short-term context aggregation, which aligns and aggregates, from neighbor frames, local regions that are closely related to the boundary context of missing regions into the target frame (The target frame refers to the current input frame under inpainting.). Furthermore, we propose dynamic long-term context aggregation to globally refine the feature map generated in the encoding stage using long-term frame features, which are dynamically updated throughout the inpainting process. Experiments show that it outperforms state-of-the-art methods with better inpainting results and fast inpainting speed. © 2020, Springer Nature Switzerland AG.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

Competing for pixels: a self-play algorithm for weakly-supervised segmentation

arXiv

引用

arXiv 2024年

作者： Saeed, Shaheer U. Huang, Shiqi Ramalhinho, Joao Gayo, Iani J.M.B. Montaña-Brown, Nina Bonmati, Ester Pereira, Stephen P. Davidson, Brian Barratt, Dean C. Clarkson, Matthew J. Hu, Yipeng Centre for Medical Image Computing Wellcome/EPSRC Centre for Interventional and Surgical Sciences Department of Medical Physics and Biomedical Engineering University College London LondonWC1E 6BT United Kingdom School of Optics and Photonics Beijing Institute of Technology Beijing China School of Computer Science and Engineering University of Westminster LondonW1W 6UW United Kingdom Institute for Liver and Digestive Health University College London LondonNW3 2QG United Kingdom Division of Surgery and Interventional Sciences University College London LondonWC1E 6BT United Kingdom

Weakly-supervised segmentation (WSS) methods, reliant on image-level labels indicating object presence, lack explicit correspondence between labels and regions of interest (ROIs), posing a significant challenge. Despite this, WSS methods have attracted attention due to their much lower annotation costs compared to fully-supervised segmentation. Leveraging reinforcement learning (RL) self-play, we propose a novel WSS method that gamifies image segmentation of a ROI. We formulate segmentation as a competition between two agents that compete to select ROI-containing patches until exhaustion of all such patches. The score at each time-step, used to compute the reward for agent training, represents likelihood of object presence within the selection, determined by an object presence detector pretrained using only image-level binary classification labels of object presence. Additionally, we propose a game termination condition that can be called by either side upon exhaustion of all ROI-containing patches, followed by the selection of a final patch from each. Upon termination, the agent is incentivised if ROIcontaining patches are exhausted or disincentivised if an ROIcontaining patch is found by the competitor. This competitive setup ensures minimisation of over- or under-segmentation, a common problem with WSS methods. Extensive experimentation across four datasets demonstrates significant performance improvements over recent state-of-the-art methods. Copyright © 2024, The Authors. All rights reserved.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Systematic Literature Review on Cyber Situational Awareness Visualizations

arXiv

引用

arXiv 2021年

作者： Jiang, Liuyue Jayatilaka, Asangi Nasim, Mehwish Grobler, Marthie Zahedi, Mansooreh Ali Babar, M. CREST - the Centre for Research on Engineering Software Technologies School of Computer Science The University of Adelaide Australia Australia College of Science and Engineering Flinders University AdelaideSA5000 Australia CSIRO's Data61 Melbourne Australia School of Computing and Information Systems The University of Melbourne Australia

The dynamics of cyber threats are increasingly complex, making it more challenging than ever for organizations to obtain in-depth insights into their cyber security status. Therefore, organizations rely on Cyber Situational Awareness (CSA) to support them in better understanding the threats and associated impacts of cyber events. Due to the heterogeneity and complexity of cyber security data, often with multidimensional attributes, sophisticated visualization techniques are needed to achieve CSA. However, there have been no previous attempts to systematically review and analyze the scientific literature on CSA visualizations. In this paper, we systematically select and review 54 publications that discuss visualizations to support CSA. We extract data from these papers to identify key stakeholders, information types, data sources, and visualization techniques. Furthermore, we analyze the level of CSA supported by the visualizations, alongside examining the maturity of the visualizations, challenges, and practices related to CSA visualizations to prepare a full analysis of the current state of CSA in an organizational context. Our results reveal certain gaps in CSA visualizations. For instance, the largest focus is on operational-level staff, and there is a clear lack of visualizations targeting other types of stakeholders such as managers, higher-level decision makers, and non-expert users. Most papers focus on threat information visualization, and there is a dearth of papers that visualize impact information, response plans, and information shared within teams. Interestingly, we find that only a few studies proposed visualizations to facilitate up to the projection level (i.e., the highest level of CSA), whereas most studies facilitated only the perception level (i.e., the lowest level of CSA). Most of the studies provide evidence of the proposed visualizations through toy examples and demonstrations, while only a few visualizations are employed in industrial practi

关键词： Visualization

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：