检索结果-内蒙古大学图书馆

The Journal of Machine Learning Research 2023年第1期24卷 8447-8521页

作者： Wenhao Li Bo Jin Xiangfeng Wang Junchi Yan Hongyuan Zha School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of Software Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai China School of Computer Science and Technology Key Laboratory of Mathematics and Engineering Applications Ministry of Education East China Normal University Shanghai China Department of Computer Science and Engineering Key Laboratory of Artificial Intelligence Ministry of Education Shanghai Jiao Tong University Shanghai China

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications due to non-interactivity between agents, the curse of dimensionality, and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. This paper proposes a exible fully decentralized actor-critic MARL framework, which can combine most of the actor-critic methods and handle large-scale general cooperative multi-agent settings. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, the proposed framework can achieve scalability and stability for the large-scale environment. This framework also reduces information transmission by the parameter sharing mechanism and novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multi-agent Particle Environment and StarCraft II show that the proposed decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods.

关键词： cooperative MARL decentralized actor-critic primal-dual method

来源：评论

学校读者我要写书评

暂无评论

Constrained Generative Sampling of 6-DoF Grasps

Constrained Generative Sampling of 6-DoF Grasps

引用

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

作者： Jens Lundell Francesco Verdoja Tran Nguyen Le Arsalan Mousavian Dieter Fox Ville Kyrki KTH Royal Institute of Technology Sweden Department of Electrical Engineering and Automation Intelligent Robotics Group School of Electrical Engineering Aalto University Finland NVIDIA Corporation USA Paul G. Allen School of Computer Science & Engineering University of Washington Seattle USA

Most state-of-the-art data-driven grasp sampling methods propose stable and collision-free grasps uniformly on the target object. For bin-picking, executing any of those reachable grasps is sufficient. However, for completing specific tasks, such as squeezing out liquid from a bottle, we want the grasp to be on a specific part of the object's body while avoiding other locations, such as the cap. This work presents a generative grasp sampling network, VCGS, capable of constrained 6-Degrees of Freedom (DoF) grasp sampling. In addition, we also curate a new dataset designed to train and evaluate methods for constrained grasping. The new dataset, called CONG, consists of over 14 million training samples of synthetically rendered point clouds and grasps at random target areas on 2889 objects. VCGS is benchmarked against GraspNet, a state-of-the-art unconstrained grasp sampler, in simulation and on a real robot. The results demonstrate that VCGS achieves a 10-15% higher grasp success rate than the baseline while being 2–3 times as sample efficient. Supplementary material is available on our project website.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Motion planning for a vigilant humanoid robot

Motion planning for a vigilant humanoid robot

引用

IEEE-RAS International Conference on Humanoid Robots

作者： Jean-Bernard Hayet Claudia Esteves Gustavo Arechavaleta Eiichi Yoshida Computer Science Group Centro de Investigación en Matemáticas Guanajuato Mexico Department of Mathematics Universidad de Guanajuato Guanajuato Mexico Robotics and Advance Manufacturing Group Centro de Investigación y de Estudios Avanzados del IPN (CINVESTAV) Saltillo Coahuila Mexico Intelligent Systems Research Institute National Institute for Advanced Industrial Science and Technology Tsukuba Ibaraki Japan

ISBN: (纸本)9781424445875

This paper presents a motion planner for a vigilant humanoid robot. In this context of surveillance, the robot task is to keep a distinctive point in the environment in sight during all of its motion. The method we propose consists of three main ingredients: (1) A motion planner for an appropriate simplified model of the walking robot, adapted to the particular needs of humanoid robots, that outputs an admissible path with local optimality properties. This path is guaranteed to satisfy the visibility constraints resulting both from the landmark and from the angular limits of the mechanical system; (2) A generic walking pattern generator that produces stable walking motions; (3) A generalized inverse-kinematics module to satisfy the remaining collisions and posture constraints, in particular the gaze direction. The effectiveness of this method is shown with several examples on the humanoid robot plataform HRP-2.

关键词： Motion planning Humanoid robots Legged locomotion Service robots Surveillance Monitoring Collision avoidance Robot sensing systems Mechanical systems Humans

来源：评论

学校读者我要写书评

暂无评论

From Alexnet to Transformers: Measuring the Non-linearity of Deep Neural Networks with Affine Optimal Transport

arXiv

引用

arXiv 2023年

作者： Bouniot, Quentin Redko, Ievgen Mallasto, Anton Laclau, Charlotte Arndt, Karol Struckmeier, Oliver Heinonen, Markus Kyrki, Ville Kaski, Samuel LTCI Télécom Paris Institut Polytechnique de Paris France Noah’s Ark Lab Paris France Smartly.io Finland Aalto University Finland Intelligent Robotics Group Finland Department of Computer Science University of Manchester United Kingdom

In the last decade, we have witnessed the introduction of several novel deep neural network (DNN) architectures exhibiting ever-increasing performance across diverse tasks. Explaining the upward trend of their performance, however, remains difficult as different DNN architectures of comparable depth and width – common factors associated with their expressive power – may exhibit a drastically different performance even when trained on the same dataset. In this paper, we introduce the concept of the non-linearity signature of DNN, the first theoretically sound solution for approximately measuring the non-linearity of deep neural networks. Built upon a score derived from closed-form optimal transport mappings, this signature provides a better understanding of the inner workings of a wide range of DNN architectures and learning paradigms, with a particular emphasis on the computer vision task. We provide extensive experimental results that highlight the practical usefulness of the proposed non-linearity signature and its potential for long-reaching implications. The code for our work is available at https://***/qbouniot/AffScoreDeep. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

A Review of Generalized Zero-Shot Learning Methods

arXiv

引用

arXiv 2020年

作者： Pourpanah, Farhad Abdar, Moloud Luo, Yuxuan Zhou, Xinlei Wang, Ran Lim, Chee Peng Wang, Xi-Zhao Jonathan Wu, Q.M. The Centre for Computer Vision and Deep Learning Department of Electrical and Computer Engineering University of Windsor WindsorONN9B 3P4 Canada Deakin University Australia The Department of Computer Science City University of Hong Kong Hong Kong The College of Mathematics and Statistics Shenzhen Key Lab. of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China The College of Computer Science and Software Engineering Guangdong Key Lab. of Intelligent Information Processing Shenzhen University Shenzhen518060 China

Generalized zero-shot learning (GZSL) aims to train a model for classifying data samples under the condition that some output classes are unknown during supervised learning. To address this challenging task, GZSL leverages semantic information of the seen (source) and unseen (target) classes to bridge the gap between both seen and unseen classes. Since its introduction, many GZSL models have been formulated. In this review paper, we present a comprehensive review on GZSL. Firstly, we provide an overview of GZSL including the problems and challenges. Then, we introduce a hierarchical categorization for the GZSL methods and discuss the representative methods in each category. In addition, we discuss the available benchmark data sets and applications of GZSL, along with a discussion on the research gaps and directions for future investigations. Copyright © 2020, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Industrial machine vision: lessons and challenges

引用

Optics News 1987年第2期13卷 12-17页

作者： Steven W. Holland Robert B. Tilove STEVEN W. HOLLAND is an assistant department head in the Computer Science Dept. of General Motors Research Laboratories (GMR). He is responsible for research projects involving computer vision robotics mobile robotics process planning knowledge systems natural language advanced programming environments parallel processing and statistics. ROBERT B. TILOVE is senior staff research scientist and project leader of the machine perception project at GMR. His research interests include geometric modeling model-based programming and control of intelligent automation systems.

No abstract available.

关键词： Integral imaging Machine vision Measurement and metrology Optical systems Sensors Statistics

来源：评论

学校读者我要写书评

暂无评论

Reactive High-level Behavior Synthesis for an Atlas Humanoid Robot

Reactive High-level Behavior Synthesis for an Atlas Humanoid...

引用

IEEE International Conference on robotics and Automation

作者： Spyros Maniatopoulos Philipp Schillinger Vitchyr Pong David C. Conner Hadas Kress-Gazit Verifiable Robotics Research Group Sibley School of Mechanical and Aerospace Engineering Cornell University Ithaca NY 14853 USA Robert Bosch GmbH Corporate Research Department for Cognitive Systems 70442 Stuttgart Germany Capable Humanitarian Robotics & Intelligent Systems Lab Department of Physics Computer Science and Engineering Christopher Newport University Newport News VA 23606 USA

ISBN: (纸本)9781467380270

In this work, we take a step towards bridging the gap between the theory of formal synthesis and its application to real-world, complex, robotic systems. In particular, we present an end-to-end approach for the automatic generation of code that implements high-level robot behaviors in a verifiably correct manner, including reaction to the possible failures of low-level actions. We start with a description of the system defined a priori. Thus, a non-expert user need only specify a high-level task. We automatically construct a formal specification, in a fragment of Linear Temporal Logic (LTL), that encodes the system's capabilities and constraints, the task, and the desired reaction to low-level failures. We then synthesize a reactive mission plan that is guaranteed to satisfy the formal specification, i.e., achieve the task's goals or correctly react to failures. Lastly, we automatically generate a state machine that instantiates the synthesized symbolic plan in software. We showcase our approach using Team ViGIR's software and Atlas humanoid robot and present lab experiments, thus demonstrating the application of formal synthesis techniques to complex robotic systems. The proposed approach has been implemented and open-sourced as a collection of Robot Operating System (ROS) packages, which are adaptable to other systems.

关键词： Atlas Humanoid Robot formal synthesis ROS

来源：评论

学校读者我要写书评

暂无评论

Constrained Generative Sampling of 6-DoF Grasps

arXiv

引用

arXiv 2023年

作者： Lundell, Jens Verdoja, Francesco Le, Tran Nguyen Mousavian, Arsalan Fox, Dieter Kyrki, Ville Intelligent Robotics Group Department of Electrical Engineering and Automation School of Electrical Engineering Aalto University Finland KTH Royal Institute of Technology Sweden NVIDIA Corporation United States Paul G. Allen School of Computer Science & Engineering University of Washington Seattle United States

Most state-of-the-art data-driven grasp sampling methods propose stable and collision-free grasps uniformly on the target object. For bin-picking, executing any of those reachable grasps is sufficient. However, for completing specific tasks, such as squeezing out liquid from a bottle, we want the grasp to be on a specific part of the object’s body while avoiding other locations, such as the cap. This work presents a generative grasp sampling network, VCGS, capable of constrained 6-Degrees of Freedom (DoF) grasp sampling. In addition, we also curate a new dataset designed to train and evaluate methods for constrained grasping. The new dataset, called CONG, consists of over 14 million training samples of synthetically rendered point clouds and grasps at random target areas on 2889 objects. VCGS is benchmarked against GraspNet, a state-of-the-art unconstrained grasp sampler, in simulation and on a real robot. The results demonstrate that VCGS achieves a 10–15% higher grasp success rate than the baseline while being 2–3 times as sample efficient. Supplementary material is available on our project website. Copyright © 2023, The Authors. All rights reserved.

关键词： Bottles

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Analysis of Skin Cancer Classification with Convolutional Neural Networks for E-Health Applications

Data-Driven Analysis of Skin Cancer Classification with Conv...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Imran Ahmed Misbah Ahmad Abdellah Chehri Gwanggil Jeon School of Computing and Information Science Anglia Ruskin University Cambridge UK Animal and Agriculture Department Hartpury University Gloucester UK Centre for Machine Vision Bristol Robotics Laboratory University of the West of England Bristol UK Department of Mathematics and Computer Science The Royal Military College of Canada (RMC) Canada Department of Embedded Systems Engineering Incheon National University Incheon Korea

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

This study explores the effectiveness of Convolutional Neural Networks (CNNs) in automatically classifying skin cancer for e-health applications. The trained model showcases impressive performance by leveraging the HAM10000 dataset, which includes a wide range of skin lesion images from seven different classes. The parameters and architecture of the CNN model are presented in a systematic manner, providing valuable insights into the reasoning behind its design. The model is optimized using the Adam optimizer and annealing techniques to ensure efficient convergence. The model’s performance is assessed on validation and test datasets, showcasing an accuracy of 78.55% and 76.49%, respectively, for skin cancer classification. This study highlights the significant potential of CNN as a powerful tool for automating the diagnosis of skin cancer, which is in line with the growing trend of using deep learning for medical image analysis.

关键词： Deep learning Systematics Image analysis Market research Skin Convolutional neural networks Lesions Global communication Medical diagnostic imaging Skin cancer

来源：评论

学校读者我要写书评

暂无评论

Integrated Bi-Manual Motion Generation and Control shaped for Probabilistic Movement Primitives

Integrated Bi-Manual Motion Generation and Control shaped fo...

引用

IEEE-RAS International Conference on Humanoid Robots

作者： Jonathan Vorndamme João Carvalho Riddhiman Laha Dorothea Koert Luis Figueredo Jan Peters Sami Haddadin Department of Computer Engineering Technische Universität München (TUM) TUM School of Computation Information and Technology (CIT) Munich Institute of Robotics and Machine Intelligence (MIRMI) Germany Intelligent Autonomous Systems group Technische Universitat Darmstadt (TUD) Germany Center for Cognitive Science TUD Germany

ISBN: (纸本)9798350309805

This work introduces a novel cooperative control framework that allows for real-time reactiveness and adaptation whilst satisfying implicit constraints stemming from proba-bilistic/stochastic trajectories. Stemming from task-oriented sampling and/or task-oriented demonstrations, e.g., learning based on motion primitives, such trajectories carry additional information often neglected during real-time control deployment. In particular, methods such as probabilistic movement primitives offer the advantage to capture the inherent stochasticity in human demonstrations - which in turn reflects human's understanding about task-variability and adaption possibilities. This information, however, is often poorly exploited and, mostly, used during offline trajectory planning stage. Our work instead introduces a novel real-time motion-generation strategy that explicitly exploits such information to improve trajectories according to changes in the environmental condition and robot task-space topology. The proposed solution is particularly well-suited for bi-manual and coordinated systems where the increased kinematic complexity, tightly-coupled constraints and reduced workspace have detrimental effects on the manipula-bility, joint-limits, and are even capable of causing unstable behavior and task-failure. Our methodology addresses these challenges, and improves performance and task-execution by taking the confidence range region explicitly into account whilst maneuvering towards better configurations. Furthermore, it can directly cope with different closed-chain kinematics and task-space topologies, resulting for instance from different grasps. Experimental evaluations on a bi-manual Franka panda robot show that the method can run in the inner control loop of the robot and enables successful execution of highly constrained tasks.

关键词： Trajectory planning Robot kinematics Humanoid robots Kinematics Probabilistic logic Real-time systems Trajectory

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：