检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Dhahri, Rayen Immer, Alexander Charpentier, Betrand Günnemann, Stephan Fortuin, Vincent Department of Computer Science Technical University of Munich Germany Munich Center for Machine Learning Germany Department of Computer Science ETH Zürich Switzerland Max Planck Institute for Intelligent Systems Tübingen Germany Pruna AI Munich Germany Helmholtz AI Munich Germany

Neural network sparsification is a promising avenue to save computational time and memory costs, especially in an age where many successful AI models are becoming too large to naïvely deploy on consumer hardware. While much work has focused on different weight pruning criteria, the overall sparsifiability of the network, i.e., its capacity to be pruned without quality loss, has often been overlooked. We present Sparsifiability via the Marginal likelihood (SpaM), a pruning framework that highlights the effectiveness of using the Bayesian marginal likelihood in conjunction with sparsity-inducing priors for making neural networks more sparsifiable. Our approach implements an automatic Occam’s razor that selects the most sparsifiable model that still explains the data well, both for structured and unstructured sparsification. In addition, we demonstrate that the pre-computed posterior precision from the Laplace approximation can be re-used to define a cheap pruning criterion, which outperforms many existing (more expensive) approaches. We demonstrate the effectiveness of our framework, especially at high sparsity levels, across a range of different neural network architectures and datasets. Copyright © 2024, The Authors. All rights reserved.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

vPlanSim: An Open Source Graphical Interface for the Visualisation and Simulation of AI systems 31

vPlanSim: An Open Source Graphical Interface for the Visuali...

引用

31st International Conference on Automated Planning and Scheduling, ICAPS 2021

作者： Roberts, Jamie O. Mastorakis, Georgios Lazaruk, Brad Franco, Santiago Stokes, Adam A. Bernardini, Sara School of Engineering Institute for Integrated Micro and Nano Systems University of Edinburgh United Kingdom EPSRC CDT in Robotics and Autonomous Systems Edinburgh Centre for Robotics United Kingdom Department of Computer Science Royal Holloway University of London United Kingdom

ISBN: (纸本)9781713832317

We introduce vPlanSim, an open source tool to aid in AI PDDL development. This tool is primarily aimed at researchers and developers who need a visual representation of their planning problem so that they can make useful insights into the performance of their system, and also to naturally convey their system to others. It is an open-source tool which allows a user to quickly and easily visualise a target environment to generate the problem files and also to visualise a plan. It is particularly well suited to spatial planning problems. This paper will demonstrate vPlanSim on 2D and 3D planning problems. vPlanSim is based on a small and carefully considered set of dependencies such as VTK and PyQt. It can be set up on different platforms and compiled from source with minimal effort. The code is and maintained via a clear code review mechanism. We welcome contributions from the open-source community. Copyright © 2021, Proceedings of the 31st International Conference on Automated Planning and Scheduling. All rights reserved.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Virtual Reality-Based Teleoperation System for Robot Forklifts

Virtual Reality-Based Teleoperation System for Robot Forklif...

引用

IEEE International Conference on Autonomous Robot systems and Competitions (ICARSC)

作者： Manuel B. Couto Marcelo R. Petry Abel Mendes Manuel F. Silva CRIIS - Centre for Robotics in Industry and Intelligent Systems INESC TEC - Institute for Systems and Computer Engineering Technology and Science Porto Portugal FEUP - Faculty of Engineering University of Porto Porto Portugal Flowbotic Coimbra Portugal ISEP - Instituto Superior de Engenharia do Porto Instituto Politécnico do Porto Porto Portugal

ISBN: (数字)9798331538606

ISBN: (纸本)9798331538613

The growing reliance on e-commerce and the demand for efficient intralogistics operations have increased the need for automation, while labour shortages continue to pose significant challenges. When combined with the inherent risks of forklift operation, this circumstance prompted businesses to look for robotic solutions for intralogistics tasks. However, robots are still limited when they come across situations that are outside of their programming scope and often need assistance from humans. To achieve the long-term goal of enhancing intralogistics operation, we propose the development of a virtual reality-based teleoperation system that allows remote operation of robot forklifts with minimal latency. Considering the specificities of the teleoperation process and network dynamics, we conduct detailed modelling to analyse latency factors, optimise system performance, and ensure a seamless user experience. Experimental results on a mobile robot have shown that the proposed teleoperation system achieves an average glass-to-glass latency of 368 ms, with capturing latency contributing to approximately 60% of the total delay. The results also indicate that network os-cillations significantly impact image quality and user experience, emphasising the importance of a stable network infrastructure.

关键词： Radio frequency Visualization 5G mobile communication System performance Robot vision systems Virtual reality Cameras User experience Wireless fidelity Testing

来源：评论

学校读者我要写书评

暂无评论

Autonomous Building of Structures in Unstructured Environments via AI Planning 31

Autonomous Building of Structures in Unstructured Environmen...

引用

31st International Conference on Automated Planning and Scheduling, ICAPS 2021

作者： Roberts, Jamie O. Franco, Santiago Stokes, Adam A. Bernardini, Sara School of Engineering Institute for Integrated Micro and Nano Systems University of Edinburgh United Kingdom EPSRC CDT in Robotics and Autonomous Systems Edinburgh Centre for Robotics United Kingdom Department of Computer Science Royal Holloway University of London United Kingdom

ISBN: (纸本)9781713832317

In this paper, we offer a novel AI planning representation, based on a Cartesian coordinate system, for enabling the autonomous operations of Multi-Robot systems in 3D environments. Each robot in the system has to conform to unique actuation and connection constraints that create a complex set of valid configurations. Our approach allows Multi-Robot systems to self-assemble themselves into larger structures via AI planning, with the overarching goal of providing structural capabilities in harsh and uncertain environments. In comparing four different PDDL (Planning Domain Definition Language) domain representations, we show that our novel formulation satisfies the practical requirements emerging from robot deployment in the real world, resulting in an AI planning system that is accurate and efficient. We scale up performance by implementing direct FDR (Finite Domain Representation) generation based on the best performing PDDL model, bypassing the PDDL-to-FDR translation used by the majority of modern planners. The proposed approach is general and can be applied to a broad range of AI problems involving reasoning in 3D spaces. Copyright © 2021, Proceedings of the 31st International Conference on Automated Planning and Scheduling. All rights reserved.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

Steps Toward Industry 5.0:Building“6S”Parallel Industries With Cyber-Physical-Social Intelligence

引用

IEEE/CAA Journal of Automatica Sinica 2023年第8期10卷 1692-1703页

作者： Xingxia Wang Jing Yang Yutong Wang Qinghai Miao Fei-Yue Wang Aijun Zhao Jian-Ling Deng Lingxi Li Xiaoxiang Na Ljubo Vlacic School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China IEEE Macao University of Science and Technology Macao 999078China Beijing Engineering Research Center of Intelligent Systems and Technology Chinese Academy of SciencesBeijing 100098 State Key Laboratory for Management and Control of Complex Systems Chinese Academy of SciencesBeijing 100190China North Automatic Control Technology Institute Taiyuan 030006China China Huaneng Group Co. Ltd.Beijing 100031China Department of Electrical and Computer Engineering Indiana University-Purdue University Indianapolis(IUPUI)IndianapolisIN 46202 USA Department of Engineering University of CambridgeCambridgeCB21PZUK School of Engineering and Built Environment Griffith UniversityGold CoastQLD 4222Australia

Very recently,intensive discussions and studies on Industry 5.0 have sprung up and caused the attention of researchers,entrepreneurs,and policymakers from various sectors around the ***,there is no consensus on why and what is Industry 5.0 *** this paper,we define Industry 5.0from its philosophical and historical origin and evolution,emphasize its new thinking on virtual-real duality and human-machine interaction,and introduce its new theory and technology based on parallel intelligence(PI),artificial societies,computational experiments,and parallel execution(the ACP method),and cyber-physical-social systems(CPSS).Case studies and applications of Industry 5.0 over the last decade have been briefly summarized and analyzed with suggestions for its future *** believe that Industry 5.0 of virtual-real interactive parallel industries has great potentials and is critical for building smart *** are outlined to ensure a roadmap that would lead to a smooth transition from CPS-based Industry 4.0 to CPSS-based Industry 5.0 for a better world which is Safe in physical spaces,S ecure in cyberspaces,Sustainable in ecology,Sensitive in individual privacy and rights,Service for all,and Smartness of all.

关键词： ACP artificial intelligence CPS CPSS Industry 4.0 Industry 5.0 parallel industries parallel intelligence

来源：评论

学校读者我要写书评

暂无评论

Predictive Modeling of Portuguese Student Performance: Comparative Machine Learning Analysis

Predictive Modeling of Portuguese Student Performance: Compa...

引用

Telecommunications Conference (ITC-Egypt), International

作者： Faris H. Rizk Mahmoud Elshabrawy Mohamed Basant Sameh Ahmed Mohamed Zaki Marwa M. Eid El-Sayed M. El-kenawy Computer Science and Intelligent Systems Research Center Blacksburg Virginia USA Department of Communications and Electronics Delta Higher Institute of Engineering and Technology Mansoura Egypt Faculty of Artificial Intelligence Delta University for Science and Technology Mansoura Egypt

ISBN: (数字)9798350351408

ISBN: (纸本)9798350351415

Such an analysis of different machine learning methods for predicting the achievement levels of students in Portuguese secondary education makes this essay. The research highlights the importance of accurate expectations of learners' results for education system administrations and respective policymakers. The current study makes use of the “Student Performance in Portuguese Secondary education” dataset and employs machine learning algorithms, namely MLPRegressor, XGBoost, DecisionTreeRegressor, CatBoost, and KNeighborsRe-gressor, to the corpus. Performance metrics such as mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), etc., are used to judge every model's performance. The conclusion that can be drawn from the data is that the MLPRegressor model leads among the competitors, having an MSE equivalent of 0.0103, which is superior to others. The findings of this study are of great significance for educational institutions and policymakers as they work to make appropriate contact with students' performance prediction.

关键词： Measurement Machine learning algorithms Education Machine learning Predictive models Market research Data models

来源：评论

学校读者我要写书评

暂无评论

Incorporating Rotation Invariance with Non-invariant Networks for Point Clouds

Incorporating Rotation Invariance with Non-invariant Network...

引用

International Conference on 3D Imaging, Modeling, Processing, Visualization and Transmission (3DIMPVT)

作者： Jiajun Fei Zhidong Deng Department of Computer Science and Technology Tsinghua University State Key Laboratory of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology (BNRist) Institute for Artificial Intelligence at Tsinghua University (THUAI) Beijing China

ISBN: (数字)9798350362459

ISBN: (纸本)9798350362466

Rotation invariance is a fundamental requirement of point cloud processing when input point clouds are not aligned. Many non-invariant networks performing well on aligned point clouds do not perform equivalent to rotated ones. Thus non-invariant and invariant networks are developed separately and only benefit a little from each other, leading to repetitive and wasteful research efforts. In this paper, we aim to bridge this gap and incorporate rotation invariance with non-invariant networks for point clouds. To this end, we propose a novel rotation invariant learning method based on efficient invariant poses (EIPs). EIPs do not rely on novel features or operations. Instead, they only rotate input point clouds into invariant poses and apply non-invariant networks in feature processing. As the name implies, EIPs have negligible complexities (efficient) and solid theoretical foundations (invariant). Experimental results demonstrate that EIPs have competitive performances on several tasks. Without using new features or operations, EIPs yield the best results on ScanObjectNN (PB T50 RS) classification and ShapeNetPart segmentation task. Our code is available at: https://***/JaronTHU/EIP.

关键词： Point cloud compression Bridges Representation learning Three-dimensional displays Codes Solids Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Enhancing Student Performance Prediction with Greylag Goose Optimization Algorithm

Enhancing Student Performance Prediction with Greylag Goose ...

引用

Telecommunications Conference (ITC-Egypt), International

ISBN: (数字)9798350351408

ISBN: (纸本)9798350351415

The current paper addresses the central role of hyperparameter optimization in improving the predictive power of the MLP Regressor for forecasting student performance in Portuguese secondary schools. The uniqueness of this research lies in its exploration of metaheuristic optimization algorithms, specifically highlighting GGO (Greylag Goose Optimization) for enhancement. The study utilized a dataset crucial for understanding and predicting student performance, with a special focus on its distinctive features. By comprehensively tuning the MLP Regressor, the paper demonstrates remarkable improvements in various performance measures, as evident in the enclosed tables. Specifically, the MSE values calculated for the MLP Regressor both before and after GGO optimization are compared. Without optimization, the MLP Regressor had an MSE of 0.0103. After GGO optimization, the MSE significantly improved to 0.0060, indicating enhanced accuracy with GGO in the model. These findings emphasize that hyperparameter optimization, particularly the GGO technique, is crucial for refining the MLP Regressor in predicting student performance. The paper not only delves into the technical aspects but also concludes by highlighting the broader consequences of these outcomes. The potential educational applications are substantial, as improved models can provide more accurate predictions, empowering educators and policymakers to make informed decisions in education. This paper establishes a foundation for future research directions, contributing to the pool of ideas for educational predictive modeling.

关键词： Accuracy Education Refining Metaheuristics Predictive models Hyperparameter optimization Prediction algorithms

来源：评论

学校读者我要写书评

暂无评论

Fine-Tuning Point Cloud Transformers with Dynamic Aggregation

Fine-Tuning Point Cloud Transformers with Dynamic Aggregatio...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Jiajun Fei Zhidong Deng Department of Computer Science and Technology Institute for Artificial Intelligence Tsinghua University (THUAI) State Key Laboratory of Intelligent Technology and Systems Beijing National Research Center for Information Science and Technology (BNRist) Tsinghua University Beijing China

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Point clouds play an important role in 3D analysis, which has broad applications in robotics and autonomous driving. The pre-training fine-tuning paradigm has shown great potential in the point cloud domain. Full fine-tuning is generally effective but leads to a heavy storage and computational burden, which becomes inefficient and unacceptable as the size of pre-trained models scales. Although efficient fine-tuning approaches have significant progress in other domains, they generally perform worse for point clouds. To overcome this dilemma, we revisit the official Point-MAE implementation and find the critical role of aggregation in fine-tuning performances. Inspired by such discoveries, we propose a novel dynamic aggregation (DA) method to replace previous static aggregation like mean or max pooling for pre-trained point cloud Transformers. Besides standard metrics such as accuracy or mIoU, we evaluate the number of tunable parameters and additional FLOPs for a fair comparison of our method to different fine-tuning approaches. We construct several DA variants and validate them through extensive experiments. Experimental results demonstrate that DA has competitive performances against full fine-tuning and other efficient fine-tuning approaches. The code is publicly available at https://***/JaronTHU/DynamicAggregation.

关键词： Point cloud compression Measurement Three-dimensional displays Codes Computational modeling Aggregates Transformers

来源：评论

学校读者我要写书评

暂无评论

Toward Fine Contact Interactions: Learning to Control Normal Contact Force with Limited Information

Toward Fine Contact Interactions: Learning to Control Normal...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Jinda Cui Jiawei Xu David Saldana Jeff Trinkle Lehigh AIRLab Honda Research Institute USA San Jose CA USA Department of Computer Science and Engineering Autonomous and Intelligent Robotics Laboratory (AIRLab) Lehigh University Bethleham PA USA

Dexterous manipulation of objects through fine control of physical contacts is essential for many important tasks of daily living. A fundamental ability underlying fine contact control is compliant control, i.e., controlling the contact forces while moving. For robots, the most widely explored approaches heavily depend on models of manipulated objects and expensive sensors to gather contact location and force information needed for real-time control. The models are difficult to obtain, and the sensors are costly, hindering personal robots' adoption in our homes and businesses. This study performs model-free reinforcement learning of a normal contact force controller on a robotic manipulation system built with a low-cost, information-poor tactile sensor. Despite the limited sensing capability, our force controller can be combined with a motion controller to enable fine contact interactions during object manipulation. Promising results are demonstrated in non-prehensile, dexterous manipulation experiments.

关键词：

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：