检索结果-内蒙古大学图书馆

AI Fairness-From Machine Learning to Federated Learning

Computer Modeling in engineering & Sciences 2024年第5期139卷 1203-1215页

作者： Lalit Mohan Patnaik Wenfeng Wang Consciousness Studies Program School of HumanitiesNational Institute of Advanced StudiesBangalore560012India Research Institute of Intelligent Engineering and Data Applications Shanghai Institute of TechnologyShanghai201418China Research Center of Ecology and Environment of Central Asia Chinese Academy of SciencesUrumqi830011China Applied Nonlinear Science Lab Anand International College of EngineeringJaipur391320India London Institute of Technology The ASE-London CTI of SCOLondonCR26EQUK Sino-Indian Joint Research Center of AI and Robotics The IMT InstituteBhubaneswar752054India

This article reviews the theory of fairness in AI-frommachine learning to federated learning,where the constraints on precision AI fairness and perspective solutions are also *** a reliable and quantitative evaluation of AI fairness,many associated concepts have been proposed,formulated and ***,the inexplicability of machine learning systems makes it almost impossible to include all necessary details in the modelling stage to ensure *** privacy worries induce the data unfairness and hence,the biases in the datasets for evaluating AI fairness are *** imbalance between algorithms’utility and humanization has further reinforced *** for federated learning systems,these constraints on precision AI fairness still *** solution is to reconcile the federated learning processes and reduce biases and imbalances accordingly.

关键词： Formulation evaluation classification constraints imbalance biases

来源：评论

学校读者我要写书评

暂无评论

Unified Editing of Panorama, 3D Scenes, and Videos Through Disentangled Self-Attention Injection

arXiv

引用

arXiv 2024年

作者： Kwon, Gihyun Park, Jangho Ye, Jong Chul Kim Jaechul Graduate School of AI KAIST Korea Republic of Department of Bio and Brain Engineering KAIST Korea Republic of Robotics Program KAIST Korea Republic of

While text-to-image models have achieved impressive capabilities in image generation and editing, their application across various modalities often necessitates training separate models. Inspired by existing method of single image editing with self attention injection and video editing with shared attention, we propose a novel unified editing framework that combines the strengths of both approaches by utilizing only a basic 2D image text-to-image (T2I) diffusion model. Specifically, we design a sampling method that facilitates editing consecutive images while maintaining semantic consistency utilizing shared self-attention features during both reference and consecutive image sampling processes. Experimental results confirm that our method enables editing across diverse modalities including 3D scenes, videos, and panorama images. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

ED-NERF: EFFICIENT TEXT-GUIDED EDITING OF 3D SCENE WITH LATENT SPACE NERF

arXiv

引用

arXiv 2023年

作者： Park, Jangho Kwon, Gihyun Ye, Jong Chul Kim Jaechul Graduate School of AI Korea Republic of Robotics Program Department of Bio and Brain Engineering KAIST Korea Republic of

Recently, there has been a significant advancement in text-to-image diffusion models, leading to groundbreaking performance in 2D image generation. These advancements have been extended to 3D models, enabling the generation of novel 3D objects from textual descriptions. This has evolved into NeRF editing methods, which allow the manipulation of existing 3D objects through textual conditioning. However, existing NeRF editing techniques have faced limitations in their performance due to slow training speeds and the use of loss functions that do not adequately consider editing. To address this, here we present a novel 3D NeRF editing approach dubbed ED-NeRF by successfully embedding real-world scenes into the latent space of the latent diffusion model (LDM) through a unique refinement layer. This approach enables us to obtain a NeRF backbone that is not only faster but also more amenable to editing compared to traditional image space NeRF editing. Furthermore, we propose an improved loss function tailored for editing by migrating the delta denoising score (DDS) distillation loss, originally used in 2D image editing to the three-dimensional domain. This novel loss function surpasses the well-known score distillation sampling (SDS) loss in terms of suitability for editing purposes. Our experimental results demonstrate that ED-NeRF achieves faster editing speed while producing improved output quality compared to state-of-the-art 3D editing models. Code and rendering results are available at our project page1 © 2023, CC BY.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Development of Detection Kidney Function Stand Alone System and Design based on Oral Ammonia Levels

Development of Detection Kidney Function Stand Alone System ...

引用

International Conference of Computer and Informatics engineering (IC2IE)

作者： William Natasha Angeline Hansen Yosica Mariana Elioenai Sitepu Winda Astuti Melinda Siswanto Automotive and Robotics Program Computer Engineering Department BINUS ASO School of Engineering Bina Nusantara University Jakarta Indonesia Product Design Program Industrial Engineering Department BINUS ASO School of Engineering Bina Nusantara University Jakarta Indonesia Industrial Engineering Department BINUS Graduate Program Master of Industrial Engineering Bina Nusantara University Jakarta Indonesia Biotechnology Engineering Department Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331505530

ISBN: (纸本)9798331505547

The development of detection kidney function system at the first prototype have problem with the dependence on the internet, which make not practical in the development of the system and the initial design was not ergonomic. In this work the development of the second step is done with the improvement of accuracy compared with other methods. In this work the new design of the embedded system is also applied and in term of algorithm, the system applied for woman patient. Three different types of machine learning techniques, namely K-nearest neighbors (KNN), Decision Tree and Random Forest are applied and compared. The AI-based kidney failure severity identification system with KNN algorithm had an average accuracy of 90% and 91% for training and testing accuracy, respectively. The design of the embedded system also renews with stylist, simple and practical.

关键词： Training Accuracy Machine learning algorithms Embedded systems Nearest neighbor methods Vectors Decision trees Kidney Random forests Testing

来源：评论

学校读者我要写书评

暂无评论

Pose Synchronization for Semi-Autonomous Dynamic Robotic Swarms with a Passivity-Short Human Operator

Pose Synchronization for Semi-Autonomous Dynamic Robotic Swa...

引用

Annual Conference of Industrial Electronics Society

作者： Reo Kanazawa Toshiyuki Murao Hiroyuki Kawai Graduate Program in Mechanical Engineering Kanazawa Institute of Technology Ishikawa Japan Department of Robotics Kanazawa Institute of Technology Ishikawa Japan

This paper presents a pose synchronization method for semi-autonomous dynamic robotic swarms with a passivityshort human operator. In the human-robot network, a human operator sends a command signal to and receives visual information from only a part of the robotic group. In this setting, the proposed dynamic control scheme achieves not only pose synchronization of all mobile robots but also convergence to the reference pose that the human operator specifies. The main advantage of the proposed method is to guarantee convergence of not only the position but also the orientation to the reference ones in the human-robot network. Finally, we provide human-in-the-loop simulation results through the human operation using a tablet PC to illustrate the performance of the proposed control method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mechanical Response of Tungsten g-TPMS Lattice Structures

SSRN

引用

SSRN 2024年

作者： Islam, Zahabul Ahmed, Tanvir Mayyas, Mohammad Abouheaf, Mohammed Mechanical and Manufacturing Engineering Program School of Engineering Bowling Green State University Bowling GreenOH43403 United States Robotics Engineering School of Engineering Bowling Green State University Bowling GreenOH43403 United States

The present study explores the mechanical behavior of tungsten gyroid Triply Periodic Minimal Surface (g-TPMS) lattice structures through atomistic simulations, focusing on models with varying lattice thickness or simply volume fractions. It examines the isotropic nature of these structures under both tensile and compressive loads. The research demonstrates that strength and ductility in g-TPMS can be adjusted through design, showcasing unique behaviors not found in bulk materials. A relationship between elastic modulus and volume fraction is established, providing coefficients for multi-scale modeling and simulation. Additionally, the study investigates the temperature-dependent mechanical response and the effects of helium embrittlement on tungsten g-TPMS and bulk tungsten, emphasizing the need for tailored materials in advanced applications like nuclear fusion reactors. © 2024, The Authors. All rights reserved.

关键词： Volume fraction

来源：评论

学校读者我要写书评

暂无评论

Stability Guaranteed Actor-Critic Learning for Robots in Continuous Time

Stability Guaranteed Actor-Critic Learning for Robots in Con...

引用

International Conference on Electrical engineering, Computing Science and Automatic Control (CCE)

作者： Luis Pantoja-Garcia Vicente Parra-Vega Rodolfo Garcia-Rodriguez Robotics and Adv. Manufacturing CINVESTAV Saltillo Mexico Postgraduate Program in Aerospacial Engineering Universidad Politecnica Metropolitana de Hidalgo Tolcayuca

Actor-Critic (AC) architecture has the salient feature, for the plethora of Reinforcement Learning schemes, that two intertwining neural networks (NN) collaborate to deploy a motor learning mechanism that oversees and evaluates the action to control a dynamical system as the robot manipulator. Since such NNs can be studied as dynamical systems to learn how to control the robot dynamics with a given performance, it has finally paved the way to deal with stability analysis, unfortunately, few works have addressed it. In this paper, we propose a Critic-NN whose approximation of the value function substantiates the decision making mechanism that collaborates to tune the Actor-NN action (approximation of inverse dynamics). The novel proposed design of the adaptation of the neural weights yields Lyapunov stability that provides explicit conditions for an attractive invariant set that render a stable regime using a quite simple NN with one hidden layer. Numerical simulations show the performance of the proposed approach. In addition, robustness is analyzed when the robot is subject to Liptchitz disturbances, interestingly showing relaunching of the learning mechanism when needed. Finally, a discussion on dealing with asymptotic stability, robustness issues, and the learning mechanism from a reward provided by the expert user is addressed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Estimation of Wrist Chattering Angle Based on Disturbance Observer 16

Estimation of Wrist Chattering Angle Based on Disturbance Ob...

引用

16th International Congress on Image and Signal Processing, BioMedical engineering and Informatics, CISP-BMEI 2023

作者： Zhang, Xinyu Sun, Zhenyu Shu, Bairui Yokoi, Hiroshi Yong, Xu Li, Guanglin Jing, Xiaobei The Siat Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518055 China The University of Electro-Communications Joint Doctoral Program for Sustainability Research Tokyo182-8585 Japan The University of Electro-Communications Department of Mechanical Engineering and Intelligent Systems Tokyo182-8585 Japan

ISBN: (纸本)9798350330755

Due to the physiological constraints and daily habits of hand movements, the wrist chattering effect (WCE) will inadvertently cause disturbance to the accuracy of fingers' pattern recognition, when healthy subjects or amputees undergoing mirror training control their fingers according to instructions. This paper proposes a method for estimating WCE angles using a disturbance observer. Based on the Denavit-Hartenberg (D-H) convention and inverse kinematics, the intervals of the WCE during finger flexion and extension are determined. The research objective is to eliminate disturbance terms containing redundant information referring to sliding mode anti-jittering methods, making subsequent machine learning models less susceptible to disturbances. © 2023 IEEE.

关键词： Inverse kinematics

来源：评论

学校读者我要写书评

暂无评论

Merge LiDAR and High-precision Digital Map for Real-time Object Detection and GNSS Heading Calibration

引用

Journal of Institute of Control, robotics and Systems 2023年第4期29卷 301-307页

作者： Sim, Hongseok Baek, Seungbeom Lee, Jaewook Moon, Heechang Industrial Convergence Interdepartmental Program Autonomous Vehicle·Intelligent Robotics Hongik University Korea Republic of Mechanical and System Design Engineering Hongik University Korea Republic of

In this study, a method was proposed for correcting the object positioning error of light detection and ranging (LiDAR) due to the roll, pitch, and yaw data error of the inertial measurement unit sensor in the global navigation satellite system (GNSS) sensor by using a high-precision digital map. A real-time obstacle map was formed using Velodyne’s three-dimensional lines, VLP-32 and VLP-16. The high-precision digital map was then merged into the LiDAR map. The map obtained by combining LiDAR and high-precision digital maps did not match because of certain errors. To eliminate these errors, a method of fitting a map with high precision on the XY plane was established, and fittable criteria were determined. Algorithm verification was performed during actual driving, and errors were removed in real time. Various autonomous driving technologies can be developed using the sensor error data preprocessing process. Copyright© ICROS 2023.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

A Real-time Affordance-based Object Pose Estimation Approach for Robotic Grasp Pose Estimation

A Real-time Affordance-based Object Pose Estimation Approach...

引用

International Conference on System Science and engineering (ICSSE)

作者： Shang-Wen Wong Yu-Chen Chiu Chi-Yi Tsai Doctoral Program in Robotics College of Engineering Tamkang University New Taipei City Taiwan Department of Electrical and Computer Engineering Tamkang University New Taipei City Taiwan

This paper proposes a pose estimation system for robot grasping based on a novel Object Affordance Detection and Segmentation (OADS) network. The proposed system consists of four modules: (1) OADS network; (2) point cloud extraction; (3) object pose estimation; (4) grasp pose estimation. Based on the OADS network, the proposed system achieves affordance-based object pose estimation results. The proposed grasp pose estimation system is evaluated on a laboratory-made dual-arm robot. Experimental results show that the proposed system achieves high detection rate and high accuracy in affordance detection and segmentation tasks, leading to a high success rate in object grasping tasks with lab-made dual-arm robot.

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：