检索结果-内蒙古大学图书馆

arXiv 2021年

作者： Zhu, Yuanheng Zhao, Dongbin Zhao, Mengchen Li, Dong State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China Huawei Noah's Ark Lab Beijing China

In single-agent Markov decision processes, an agent can optimize its policy based on the interaction with environment. In multi-player Markov games (MGs), however, the interaction is non-stationary due to the behaviors of other players, so the agent has no fixed optimization objective. In this paper, we treat the evolution of player policies as a dynamical process and propose a novel learning scheme for Nash equilibrium. The core is to evolve one's policy according to not just its current in-game performance, but an aggregation of its performance over history. We show that for a variety of MGs, players in our learning scheme will provably converge to a point that is an approximation to Nash equilibrium. Combined with neural networks, we develop the empirical policy optimization algorithm, that is implemented in a reinforcement-learning framework and runs in a distributed way, with each player optimizing its policy based on own observations. We use two numerical examples to validate the convergence property on small-scale MGs with n ≥ 2 players, and a pong example to show the potential of our algorithm on large games. © 2021, CC BY-NC-ND.

关键词： Computation theory

来源：评论

学校读者我要写书评

暂无评论

Action Recognition Algorithm based on 2D Human Pose Estimation Method

Action Recognition Algorithm based on 2D Human Pose Estimati...

引用

第40届中国控制会议

作者： Chongkai Yu Wenjie Chen Ye Li Chen Chen State Key Laboratory of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing Institute of Technology Chongqing Innovation Center

This paper presents an action recognition method based on 2D human body node data in video. This method uses the pose estimation algorithm to detect the human body node data in each frame of video information. We get the two-dimensional coordinates and confidence data of the nodes, and optimize the arrangement of these data into a 3D array form similar to ***, we use the classical two-dimensional convolutional neural network to carry out classification training. The test on UCF-101 data set shows that this method can indeed improve the accuracy of action recognition based on RGB information to a certain extent, and reduce the training cost.

关键词： Action recognition Pose estimation Two-dimensional convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Parallel Vision and Learning for Intelligent Perception in Smart Driving

Parallel Vision and Learning for Intelligent Perception in S...

引用

Digital Twins and Parallel Intelligence (DTPI), IEEE International Conference on

作者： Hui Zhang Yidong Li Xuan Li Fei-Yue Wang Beijing Jiaotong University Beijing China Peng Cheng Laboratory Shenzhen China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

Modern vision models are built upon large-scale diversified dataset, which is labor-intensive and cost-ineffective to collect and label. They are usually effective under simple constrained conditions, while the performance gets worse under actual traffic scenes. To address aforementioned problems, we propose a visual analysis framework based on parallel vision and learning for intelligent perception in smart driving. Specifically, our framework first generate and select data exploiting artificial image system. Then, computational experiments are conducted with predictive learning for model design, learning, and evaluation. Our framework is powered with virtual-actual interactive ability through parallel execution, integrating information from different scenes. The difficulty of the visual model in the actual scene can be applied to guide the model training in the artificial scene. We also carried out a case study to preliminarily demonstrate the effectiveness of the proposed framework.

关键词： Training Learning systems Visualization Computational modeling Digital twin Conferences Learning (artificial intelligence)

来源：评论

学校读者我要写书评

暂无评论

An Introduction to Domain Adaptive Object Detection from Synthesis to Reality

An Introduction to Domain Adaptive Object Detection from Syn...

引用

第40届中国控制会议

作者： Zhijun Xue Wenjie Chen Jing Li State Key Laboratory of Intelligent Control and Decision of Complex Systems School of AutomationBeijing Institute of Technology Beijing Institute of Technology Chongqing Innovation Center

In recent years,deep learning based object detection has achieved great *** methods typically assume that large amount of labeled training data is available,meanwhile,training and test data are independent and identically ***,the two assumptions are not always hold in *** many applications,it is prohibitively expensive and time-consuming to obtain large quantities of labeled *** computer graphics technology to generate a large number of labeled data provides a solution to this ***,direct transfer across domains from synthesis to reality often performs poorly due to the presence of domain *** adaptive object detection are concerned with accounting for these types of *** this paper,we present an introduction to these ***,we briefly introduce the object detection and domain ***,the synthetic object detection datasets and related software tools are ***,we present a categorization of approaches,divided into discrepancy-based methods,adversarial discriminative methods,reconstruction-based methods and ***,we also discuss some potential deficiencies of current methods and several open problems which can be explored in future work.

关键词： Deep learning Object detection Synthesis to reality Domain adaptation

来源：评论

学校读者我要写书评

暂无评论

Procedure Recognition by Knowledge-Driven Segmentation in Robotic-Assisted Vitreoretinal Surgery

Procedure Recognition by Knowledge-Driven Segmentation in Ro...

引用

IEEE International Conference on Robotics and automation (ICRA)

作者： Zhen Li Yawen Deng Qiang Ye Weihong Yu Haoxiang Qi Yaliang Liu Zhangguo Yu Gui-Bin Bian School of Electronic and Information Engineering Tongji University Shanghai China Institute of Automation Chinese Academy of Sciences Beijing China School of Mechatronic Engineering Beijing Institute of Technology Beijing China Department of Ophthalmology Peking Union Medical College Hospital Chinese Academy of Medical Sciences and Peking Union Medical College Beijing China Key Laboratory of Biomimetic Robots and Systems Ministry of Education Beijing China International Joint Research Laboratory of Biomimetic Robots and Systems Ministry of Education Beijing China State Key Laboratory of Intelligent Control and Decision of Complex System Beijing China

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

Internal limiting membrane (ILM) peeling is a vital vitreoretinal surgery procedure. However, due to the thickness of just 1-2 micrometers and the intricacies associated with its varying density and adhesion, the difficulty of manipulation exceeds the physiological limits of human perception and operation. Surgical robot is characterized by high precision and stability. However, navigating intricate intraocular environments and handling minuscule high-precision areas remain enormous challenges. These include issues of uneven lighting, field-of-view loss, and motion blur. This paper proposed a perception method named ‘Multimodal Surgical Process Recognition based on Domain Knowledge and Segmentation (MSPR-DKS),’ designed to address these challenges and provide input for the precise control of robots. Moreover, a comprehensive dataset focused on ILM peeling during macular hole surgeries was established. Experimental results underscore the efficacy of this approach, with segmentation accuracies exceeding 99.27% for instruments and macular holes and an average accuracy of 98.97% in recognizing surgical processes. This study paves the way for leveraging domain knowledge and image segmentation to improve robot-assisted manipulation of soft tissues in ophthalmology.

关键词： Image segmentation Visualization Accuracy Medical robotics Motion segmentation Surgery Lighting

来源：评论

学校读者我要写书评

暂无评论

Geometry Problem Solving Based on Counter-Factual Evolutionary Reasoning

Geometry Problem Solving Based on Counter-Factual Evolutiona...

引用

IEEE International Conference on automation Science and Engineering (CASE)

作者： Bing Song Gang Xiong Zhen Shen Fenghua Zhu Yisheng Lv Peijun Ye The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China The Beijing Engineering Research Center of Intelligent Systems and Technology Institute of Automation Chinese Academy of Sciences China The Guangdong Engineering Research Center of 3D Printing and Intelligent Manufacturing The Cloud Computing Center Chinese Academy of Sciences Dongguan China Intelligent Manufacturing Center Qingdao Academy of Intelligent Industries Qingdao China

As a representative topic in natural language processing and automated theorem proving, geometry problem solving requires an abstract problem understanding and symbolic reasoning. A major challenge here is to find a feasible reasoning sequence that is consistent with given axioms and the theorems already proved. Most recent methods have exploited neural network-based techniques to automatically discover eligible solving steps. Such a kind of methods, however, is greatly impacted by the expert solutions for training. To improve the accuracy, this paper proposes a new method called counterfactual evolutionary reasoning, which uses a generative adversarial network to generate initial reasoning sequences and then introduces counterfactual reasoning to explore potential solutions. By directly exploring theorem candidates rather than the neural network selection, the new method can sufficiently extend the searching space to get a more appropriate reasoning step. Through comparative experiments on the recent proposed Geometry3k, the largest geometry problem solving dataset, our method generally achieves a higher accuracy than most previous methods, bringing an overall improvement about 4.4% compared with the transformer models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Aerial battlefield targets grouping based on DTW-DBSCAN algorithm

Aerial battlefield targets grouping based on DTW-DBSCAN algo...

引用

第40届中国控制会议

作者： Shao Zhuang Chen Chen School of Automation Beijing Institute of Technology State Key Laboratory of Intelligent Control and Decision of Complex System

Targets grouping is the basis of second-level information fusion,which can effectively assist commanders to make *** aerial targets grouping algorithm only takes the radar acquisition data at the current time as the object and cannot update the clustering results automatically.A grouping method combining dynamic time warping(DTW) and the algorithm Density-Based Spatial Clustering of Applications with Noise(DBSCAN) is *** DTW distance of each attribute historical time-series data is used to measure the similarity between ***,an improved DBSCAN algorithm is used for *** simulation results show that the method has better grouping effect and can automatically cluster regularly.

关键词： targets grouping dynamic time warping density clustering

来源：评论

学校读者我要写书评

暂无评论

Traffic-Condition-Awareness Ensemble Learning for Traffic Flow Prediction ⁎

引用

IFAC-PapersOnLine 2020年第5期53卷 582-587页

作者： Yuanyuan Chen Yisheng Lv Peijun Ye Fenghua Zhu State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing 100190 China

Traffic prediction is an elemental function of Intelligent Transportation systems (ITS), and accurate and timely prediction is significant for both proactive traffic control and providing traveler information. In this paper, we focus on investigating ensemble leaning that benefits from different base models, and propose a traffic-condition-awareness ensemble approach. We apply graph convolution on the network of traffic detectors to capture the spatial patterns embedded in traffic flow. Then, the extracted features are used to formulate a weight matrix to ensemble the predictions of base models according to their performances under a certain condition. We performed a series of experiments on a real dataset to compare the proposed methods with several competitive models, including ensemble methods: Weight Regression model and Gradient Boosting Regression Tree model, and single model approach: Support Vector Regression (SVR), Long Short-term Memory (LSTM) model and Historical Average Model model. Experimental results demonstrate that our method can significantly improve the performances of traffic flow prediction.

关键词： Intelligent road transportation Urban mobility

来源：评论

学校读者我要写书评

暂无评论

Modeling and Analysis of Mtj Array with Shielding Layer

SSRN

引用

SSRN 2023年

作者： Qu, Ying-Hui Xiong, Jing-Ya Zuo, Chao Dong, Kai-Feng Song, Jun-Lei Mo, Wen-Qin Hui, Ya-Juan Jin, Fang School of Automation China University of Geosciences Wuhan430074 China Hubei key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Geological Exploration Intelligent Technology Engineering Research Center The Ministry of Education Wuhan Second Ship Design and Research Institute Hubei Province Wuhan China

In STT-MRAM, data is stored in a magnetic tunnel structure (MTJ) containing multiple ferromagnetic layers. Each ferromagnetic layer generates a stray field, which affects the critical switching current, switching time, thermal stability, and the density of the MTJ array. These factors have a direct impact on the performance and storage capacity of STT-MRAM. Previously, we have studied that hexagonal MTJ arrays can improve storage density. However, this improvement also leads to increased intracellular and intercellular stray fields, which in turn affects the device's performance. This paper aims to reduce the intercellular stray field by covering each MTJ unit with a soft magnetic shielding layer and investigates the performance of the MTJ array with a shielding layer. Firstly, a simulation model of a single MTJ stack is built using the simulation software COMSOL. Then, based on modeling the MTJ stack, square and hexagonal MTJ arrays are established. The intercellular magnetic coupling coefficient (Ψ) is quoted, and Ψ=2% is defined as the threshold below which the intercellular magnetic coupling can be considered negligible. The results indicate that the soft magnetic shielding layer can effectively shield the stray field, reducing its impact on MTJ performance. Specifically, a thicker shielding layer with higher magnetic permeability provides better shielding efficacy. However, comparing square and hexagonal MTJ array after magnetic shielding reveals a larger stray field between the hexagonal array. © 2023, The Authors. All rights reserved.

关键词： Ferromagnetic materials

来源：评论

学校读者我要写书评

暂无评论

A Multi-Model Fusion Approach for Transformer Fault Detection

A Multi-Model Fusion Approach for Transformer Fault Detectio...

引用

Information Technology,Big Data and Artificial Intelligence (ICIBA), International Conference on IEEE International Conference on

作者： Ke Zhou Wei Zhang Ligang Li Shaoming Pan Chengwei Huang Guangxi Power Grid Equipment Monitoring and Diagnosis Technology Innovation Center Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment Electric Power Research Institute of Guangxi Power Grid CO. LTD. Nanning State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Key Laboratory of Networked Control Systems Chinese Academy of Sciences China Power Supply Bureau of Guangxi Hezhou Power Grid Co. Ltd.

ISBN: (数字)9798350363609

ISBN: (纸本)9798350363616

Addressing the weak generalization capability and suboptimal prediction performance of single mechanistic and data-driven models, this paper proposes a transformer fault detection method based on the fusion of mechanism and data models. Firstly, vibration and current data are collected on-site to train the SVM vibration model and the Adaboost current model. Secondly, current simulation data is obtained using laboratory simulation software. This simulation data, combined with the current data collected under the same conditions on-site, is used as input to train the GBDT error compensation model. Finally, these three models are used as inputs to the GRNN, completing the overall model design. This fusion of mechanistic and data-driven models not only addresses the parameter mismatch issue of the mechanistic model under complex working conditions but also deeply mines time-series data. Experimental results show that The mean squared error of the fusion model is 0.002 lower than that of the data model, and the coefficient of determination is improved by 5.07% compared to the mechanistic model, enhancing the overall predictive accuracy of the model.

关键词： Vibrations Support vector machines Training Accuracy Biological system modeling Fault detection Predictive models Transformers Prediction algorithms Data models

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：