检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Novin, Roya Sabbagh Taylor, Ellen Hermans, Tucker Merryweather, Andrew Department of Mechanical Engineering and Robotics Center University of Utah United States Center for Health Design ConcordCA United States School of Computing and Robotics Center University of Utah United States

Objectives: This study proposes a computational model to evaluate patient room design layout and features that contribute to patient stability and mitigate the risk of fall. Background: While common fall risk assessment tools in nursing have an acceptable level of sensitivity and specificity, they focus on intrinsic factors and medications, making risk assessment limited in terms of how the physical environment contributes to fall risk. Methods: We use literature to inform a computational model (algorithm) to define the relationship between these factors and the risk of fall. We use a trajectory optimization approach for patient motion prediction. Results: Based on available data, the algorithm includes static factors of lighting, flooring, supportive objects, and bathroom doors, and dynamic factors of patient movement. This preliminary model was tested using four room designs as examples of typical room configurations. Results show the capabilities of the proposed model to identify the risk associated with different room layouts and features. Conclusions: This innovative approach to room design evaluation and resulting estimation of patient fall risk show promise as proactive evidence-based tool to evaluate the relationship of potential fall risk and room design. The development of the model highlights the challenge of heterogeneity in factors and reporting found in the studies of patient falls, which hinder our understanding of the role of the built environment in mitigating risk. A more comprehensive investigation comparing the model with actual patient falls data is needed to further refine model development. Copyright © 2020, The Authors. All rights reserved.

关键词： Risk assessment

来源：评论

学校读者我要写书评

暂无评论

From Monocular Vision to Autonomous Action: Guiding Tumor Resection via 3D Reconstruction

arXiv

引用

arXiv 2025年

作者： Acar, Ayberk Smith, Mariana Al-Zogbi, Lidia Watts, Tanner Li, Fangjie Li, Hao Yilmaz, Nural Scheikl, Paul Maria d’Almeida, Jesse F. Sharma, Susheela Branscombe, Lauren Ertop, Tayfun Efe Webster, Robert J. Oguz, Ipek Kuntz, Alan Krieger, Axel Wu, Jie Ying Department of Computer Science Vanderbilt University NashvilleTN37235 United States Department of Mechanical Engineering Johns Hopkins University BaltimoreMD21211 United States Robotics Center Kahlert School of Computing University of Utah Salt Lake CityUT84112 United States Department of Mechanical Engineering Vanderbilt University NashvilleTN37235 United States Virtuoso Surgical NashvilleTN37205 United States Department of Mechanical Aerospace and Biomedical Engineering University of Tennessee KnoxvilleTN37996 United States

Surgical automation requires precise guidance and understanding of the scene. Current methods in the literature rely on bulky depth cameras to create maps of the anatomy, however this does not translate well to space-limited clinical applications. Monocular cameras are small and allow minimally invasive surgeries in tight spaces but additional processing is required to generate 3D scene understanding. We propose a 3D mapping pipeline that uses only RGB images to create segmented point clouds of the target anatomy. To ensure the most precise reconstruction, we compare different structure from motion algorithms’ performance on mapping the central airway obstructions, and test the pipeline on a downstream task of tumor resection. In several metrics, including post-procedure tissue model evaluation, our pipeline performs comparably to RGB-D cameras and, in some cases, even surpasses their performance. These promising results demonstrate that automation guidance can be achieved in minimally invasive procedures with monocular cameras. This study is a step toward the complete autonomy of surgical robots. © 2025, CC BY-NC-ND.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Catastrophic Interference in Reinforcement Learning: A Solution Based on Context Division and Knowledge Distillation

arXiv

引用

arXiv 2021年

作者： Zhang, Tiantian Wang, Xueqian Liang, Bin Yuan, Bo The Intelligent Computing Lab Shenzhen International Graduate School Tsinghua University Shenzhen518055 China The Center for Artificial Intelligence and Robotics Shenzhen International Graduate School Tsinghua University Shenzhen518055 China The Research Center for Navigation and Control Department of Automation Tsinghua University Beijing100084 China

The powerful learning ability of deep neural networks enables reinforcement learning agents to learn competent control policies directly from continuous environments. In theory, to achieve stable performance, neural networks assume i.i.d. inputs, which unfortunately does no hold in the general reinforcement learning paradigm where the training data is temporally correlated and non-stationary. This issue may lead to the phenomenon of "catastrophic interference" and the collapse in performance. In this paper, we present IQ, i.e., interference-aware deep Q-learning, to mitigate catastrophic interference in single-task deep reinforcement learning. Specifically, we resort to online clustering to achieve on-the-fly context division, together with a multi-head network and a knowledge distillation regularization term for preserving the policy of learned contexts. Built upon deep Q networks, IQ consistently boosts the stability and performance when compared to existing methods, verified with extensive experiments on classic control and Atari tasks. The code is publicly available at: https://***/Sweety-dm/Interference-aware-Deep-Q-learning. © 2021, CC BY-NC-SA.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

In-hand object-dynamics inference using tactile fingertips

arXiv

引用

arXiv 2020年

作者： Sundaralingam, Balakumar Hermans, Tucker University of Utah Robotics Center School of Computing University of Utah Salt Lake CityUT United States

Having the ability to estimate an object’s properties through interaction will enable robots to manipulate novel objects. Object’s dynamics, specifically the friction and inertial parameters have only been estimated in a lab environment with precise and often external sensing. Could we infer an object’s dynamics in the wild with only the robot’s sensors? In this paper, we explore the estimation of dynamics of a grasped object in motion, with tactile force sensing at multiple fingertips. Our estimation approach does not rely on torque sensing to estimate the dynamics. To estimate friction, we develop a control scheme to actively interact with the object until slip is detected. To robustly perform the inertial estimation, we setup a factor graph that fuses all our sensor measurements on physically consistent manifolds and perform inference. We show that tactile fingertips enable in-hand dynamics estimation of low mass objects. Copyright © 2020, The Authors. All rights reserved.

关键词： Friction

来源：评论

学校读者我要写书评

暂无评论

Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methods

引用

IEEE Transactions on Neural Networks and Learning Systems 2024年第6期36卷 9737-9757页

作者： Yuji Cao Huan Zhao Yuheng Cheng Ting Shu Yue Chen Guolong Liu Gaoqi Liang Junhua Zhao Jinyue Yan Yun Li Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Hong Kong SAR China Department of Building Environment and Energy Engineering The Hong Kong Polytechnic University Hong Kong China School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Center for Crowd Intelligence Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China School of Electrical and Electronic Engineering Nanyang Technological University Jurong West Singapore School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen China i4AI Ltd. London U.K.

With extensive pretrained knowledge and high-level general capabilities, large language models (LLMs) emerge as a promising avenue to augment reinforcement learning (RL) in aspects, such as multitask learning, sample efficiency, and high-level task planning. In this survey, we provide a comprehensive review of the existing literature in LLM-enhanced RL and summarize its characteristics compared with conventional RL methods, aiming to clarify the research scope and directions for future studies. Utilizing the classical agent-environment interaction paradigm, we propose a structured taxonomy to systematically categorize LLMs’ functionalities in RL, including four roles: information processor, reward designer, decision-maker, and generator. For each role, we summarize the methodologies, analyze the specific RL challenges that are mitigated and provide insights into future directions. Finally, the comparative analysis of each role, potential applications, prospective opportunities, and challenges of the LLM-enhanced RL are discussed. By proposing this taxonomy, we aim to provide a framework for researchers to effectively leverage LLMs in the RL field, potentially accelerating RL applications in complex applications, such as robotics, autonomous driving, and energy systems.

关键词： Taxonomy Surveys Visualization Reviews Planning Reinforcement learning Games Transformers Natural language processing Large language models

来源：评论

学校读者我要写书评

暂无评论

Bipartite Tracking Consensus for General Linear Multi-Agent Systems with Asynchronous Communications over Signed Networks

Bipartite Tracking Consensus for General Linear Multi-Agent ...

引用

第40届中国控制会议

作者： Runyao Chen Lulu Chen Jinliang Shao Wei Xing Zheng School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China School of Automation Engineering University of Electronic Science and Technology of China Research Center on Crowd Spectrum Intelligence Shenzhen Institute of Artifcial Intelligence and Robotics for Society School of Computing Engineering and Mathematics Western Sydney University

In this paper, the bipartite tracking consensus problem of general linear multi-agent systems over signed networks is investigated, in which an asynchronous communication manner is adopted. The asynchronous setting refers to that the clocks of each agent, at which the agent’s dynamics is updated, are independent of others. Relying on some related properties of the sub-stochastic matrix, the dynamics of the agents under the proposed asynchronous consensus protocol is rigorously *** is shown that the bipartite tracking consensus is implemented if the signed digraph contains a spanning tree. A simulation example is provided to verify our theoretical findings.

关键词： Bipartite tracking consensus Asynchronous communications General linear multi-agent systems Signed networks

来源：评论

学校读者我要写书评

暂无评论

Catastrophic interference in reinforcement learning: A solution based on context division and knowledge distillation

TechRxiv

引用

TechRxiv 2021年

作者： Zhang, Tiantian Wang, Xueqian Liang, Bin Yuan, Bo The Intelligent Computing Lab Division of Informatics Shenzhen International Graduate School Tsinghua University Shenzhen518055 China The Center for Artificial Intelligence and Robotics Shenzhen International Graduate School Tsinghua University Shenzhen518055 China The Research Center for Navigation and Control Department of Automation Tsinghua University Beijing100084 China

The powerful learning ability of deep neural networks enables reinforcement learning (RL) agents to learn competent control policies directly from high-dimensional and continuous environments. In theory, to achieve stable performance, neural networks assume i.i.d. inputs, which unfortunately does no hold in the general RL paradigm where the training data is temporally correlated and non-stationary. This issue may lead to the phenomenon of "catastrophic interference" (a.k.a. "catastrophic forgetting") and the collapse in performance as later training is likely to overwrite and interfer with previously learned good policies. In this paper, we introduce the concept of "context" into the single-task RL and develop a novel scheme, termed as Context Division and Knowledge Distillation (CDaKD) driven RL, to divide all states experienced during training into a series of contexts. Its motivation is to mitigate the challenge of aforementioned catastrophic interference in deep RL, thereby improving the stability and plasticity of RL models. At the heart of CDaKD is a value function, parameterized by a neural network feature extractor shared across all contexts, and a set of output heads, each specializing on an individual context. In CDaKD, we exploit online clustering to achieve context division, and interference is further alleviated by a knowledge distillation regularization term on the output layers for learned contexts. In addition, to effectively obtain the context division in high-dimensional state spaces (e.g., image inputs), we perform clustering in the lower-dimensional representation space of a randomly initialized convolutional encoder, which is fixed throughout training. Our results show that, with various replay memory capacities, CDaKD can consistently improve the performance of existing RL algorithms on classic OpenAI Gym tasks and the more complex high-dimensional Atari tasks, incurring only moderate computational overhead. © 2021, CC BY.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Multi-Fingered Active Grasp Learning

Multi-Fingered Active Grasp Learning

引用

2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Qingkai Lu Mark Van der Merwe Tucker Hermans School of Computing and the Robotics Center University of Utah Salt Lake City UT USA NVIDIA Seattle WA USA

ISBN: (数字)9781728162126

ISBN: (纸本)9781728162133

Learning-based approaches to grasp planning are preferred over analytical methods due to their ability to better generalize to new, partially observed objects. However, data collection remains one of the biggest bottlenecks for grasp learning methods, particularly for multi-fingered hands. The relatively high dimensional configuration space of the hands coupled with the diversity of objects common in daily life requires a significant number of samples to produce robust and confident grasp success classifiers. In this paper, we present the first active deep learning approach to grasping that searches over the grasp configuration space and classifier confidence in a unified manner. We base our approach on recent success in planning multi-fingered grasps as probabilistic inference with a learned neural network likelihood function. We embed this within a multi-armed bandit formulation of sample selection. We show that our active grasp learning approach uses fewer training samples to produce grasp success rates comparable with the passive supervised learning method trained with grasping data generated by an analytical planner. We additionally show that grasps generated by the active learner have greater qualitative and quantitative diversity in shape.

关键词： Training Shape Supervised learning Neural networks Grasping Probabilistic logic Planning

来源：评论

学校读者我要写书评

暂无评论

Open-Pose 3D Zero-Shot Learning: Benchmark and Challenges

arXiv

引用

arXiv 2023年

作者： Zhao, Weiguang Yang, Guanyu Zhang, Rui Jiang, Chenru Yang, Chaolong Yan, Yuyao Hussain, Amir Huang, Kaizhu Department of Computer Science University of Liverpool LiverpoolL69 7ZX United Kingdom Department of Foundational Mathematics Xi’an Jiaotong-Liverpool University Suzhou215123 China Department of Mechatronics and Robotics Xi’an Jiaotong-Liverpool University Suzhou215123 China School of Robotic Xi’an Jiaotong-Liverpool University Suzhou215123 China School of Computing Edinburgh Napier University EdinburghEH11 4BN United Kingdom Data Science Research Center Duke Kunshan University Kunshan215316 China

With the explosive 3D data growth, the urgency of utilizing zero-shot learning to facilitate data labeling becomes evident. Recently, methods transferring language or language-image pre-training models like Contrastive Language-Image Pre-training (CLIP) to 3D vision have made significant progress in the 3D zero-shot classification task. These methods primarily focus on 3D object classification with an aligned pose;such a setting is, however, rather restrictive, which overlooks the recognition of 3D objects with open poses typically encountered in real-world scenarios, such as an overturned chair or a lying teddy bear. To this end, we propose a more realistic and challenging scenario named open-pose 3D zero-shot classification, focusing on the recognition of 3D objects regardless of their orientation. First, we revisit the current research on 3D zero-shot classification, and propose two benchmark datasets specifically designed for the open-pose setting. We empirically validate many of the most popular methods in the proposed open-pose benchmark. Our investigations reveal that most current 3D zero-shot classification models suffer from poor performance, indicating a substantial exploration room towards the new direction. Furthermore, we study a concise pipeline with an iterative angle refinement mechanism that automatically optimizes one ideal angle to classify these open-pose 3D objects. In particular, to make validation more compelling and not just limited to existing CLIP-based methods, we also pioneer the exploration of knowledge transfer based on Diffusion models. While the proposed solutions can serve as a new benchmark for open-pose 3D zero-shot classification, we discuss the complexities and challenges of this scenario that remain for further research development. The code is available publicly at https://***/weiguangzhao/Diff-OP3D. © 2023, CC BY-NC-ND.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Fair Evaluation of Federated Learning Algorithms for Automated Breast Density Classification: The Results of the 2022 ACR-NCI-NVIDIA Federated Learning Challenge

arXiv

引用

arXiv 2024年

作者： Schmidt, Kendall Bearce, Benjamin Chang, Ken Coombs, Laura Farahani, Keyvan Elbatel, Marawan Mouheb, Kaouther Marti, Robert Zhang, Ruipeng Zhang, Yao Wang, Yanfeng Hu, Yaojun Ying, Haochao Xu, Yuyang Testagrose, Conrad Demirer, Mutlu Gupta, Vikash Akünal, Ünal Bujotzek, Markus Maier-Hein, Klaus H. Qin, Yi Li, Xiaomeng Kalpathy-Cramer, Jayashree Roth, Holger R. American College of Radiology United States The Massachusetts General Hospital United States University of Colorado United States National Institutes of Health National Cancer Institute United States Computer Vision and Robotics Institute University of Girona Spain Cooperative Medianet Innovation Center Shanghai Jiao Tong University China Shanghai AI Laboratory China Real Doctor AI Research Centre Zhejiang University China School of Public Health Zhejiang University China College of Computer Science and Technology Zhejiang University China University of North Florida College of Computing Jacksonville United States Mayo Clinic Florida Radiology United States Division of Medical Image Computing German Cancer Research Center Heidelberg Germany Electronic and Computer Engineering Hong Kong University of Science and Technology China NVIDIA United States

The correct interpretation of breast density is important in the assessment of breast cancer risk. AI has been shown capable of accurately predicting breast density, however, due to the differences in imaging characteristics across mammography systems, models built using data from one system do not generalize well to other systems. Though federated learning (FL) has emerged as a way to improve the generalizability of AI without the need to share data, the best way to preserve features from all training data during FL is an active area of research. To explore FL methodology, the breast density classification FL challenge was hosted in partnership with the American College of Radiology, Harvard Medical schools' Mass General Brigham, University of Colorado, NVIDIA, and the National Institutes of Health National Cancer Institute. Challenge participants were able to submit docker containers capable of implementing FL on three simulated medical facilities, each containing a unique large mammography dataset. The breast density FL challenge ran from June 15 to September 5, 2022, attracting seven finalists from around the world. The winning FL submission reached a linear kappa score of 0.653 on the challenge test data and 0.413 on an external testing dataset, scoring comparably to a model trained on the same data in a central location. © 2024, CC BY.

关键词： Learning algorithms

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：