检索结果-内蒙古大学图书馆

International Conference on Pattern Recognition

作者： Yinlin Li Yang Qian Xu Yang Yuren Zhang State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China ByteDance Beijing China

This paper presents a weakly supervised object detection method based on activity label and relationship modeling, which is motivated by the assumption that configuration of human and object are similar in same activity, and joint modeling of human, active object and activity could leverage the recognition of them. Compared to most weakly supervised method taking object as independent instance, firstly, active human and object proposals are learned and filtered based on class activation map of multi-label classification. Secondly, a spatial relationship prior including relative position, scale, overlaps etc are learned dependent on action. Finally, a multi-stream object detection framework integrating the spatial prior and pairwise ROI pooling are proposed to jointly learn the object and action class. Experiments are conducted on HICO-DET dataset, and our approach outperforms the state of the art weakly supervised object detection methods.

关键词： Training Object detection Gaussian distribution Pattern recognition Proposals

来源：评论

学校读者我要写书评

暂无评论

A parallel framework for scene understanding 1

A parallel framework for scene understanding

引用

1st IEEE International Conference on Digital Twins and Parallel Intelligence, DTPI 2021

作者： Zhang, Wenwen Wang, Jiangong Wang, Kunfeng School of Software Engineering Xi'An Jiaotong University Xi'an710049 China Institute of Automation Chinese Academy of Sciences The State Key Laboratory for Management and Control of Complex Systems Beijing100190 China College of Information Science and Technology Beijing University of Chemical Technology Beijing100029 China

ISBN: (纸本)9781665433372

Different related computer vision tasks make it possible to understand and analyze the scene totally. However, there is no useful framework combining different tasks for scene understanding. In this paper, we investigate the current state of scene understanding from different aspects relying on different computer vision tasks and propose an effective framework based on the theory of parallel vision to combine different tasks for scene understanding. With the maintained artificial scene throughout time from the proposed framework, we can understand the scene in an elegant way. To explain our proposed framework, a case study is given. © 2021 IEEE.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Embed Trajectory Imitation in Reinforcement Learning: A Hybrid Method for Autonomous Vehicle Planning

Embed Trajectory Imitation in Reinforcement Learning: A Hybr...

引用

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

作者： Yuxiao Wang Xingyuan Dai Kara Wang Hub Ali Fenghua Zhu State Key Laboratory of Management and Control for Complex System Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China School of Artificial Intelligence Anhui University Hefei China

Learning-based autonomous vehicle trajectory planning methods have shown excellent performance in a variety of complex traffic scenarios. However, the existing imitation learning (IL) and reinforcement learning (RL) algorithms still have their limitations, such as poor safety and generalizability for IL, and low data efficiency for RL. To leverage their respective advantages and mitigate the limitations, this paper proposes a novel hybrid RL algorithm for autonomous vehicle planning, where IL is embedded in it to guide its exploration with expert knowledge. Different from existing approaches, we use multi-step trajectory prediction instead of behavior cloning as the IL method integrated with online RL. Through such design, we make a further step in the research about how expert demonstration can be helpful to RL. Moreover, we conduct parallel training and testing of the algorithm based on real-world driving data. Experimental results demonstrate that our proposed approach outperforms standalone IL and RL methods, and performs better than RL methods enhanced by behavior cloning.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Heterogeneous Graph Reinforcement Learning for Dependency-aware Multi-task Allocation in Spatial Crowdsourcing

arXiv

引用

arXiv 2024年

作者： Zhao, Yong Zhu, Zhengqiu Gao, Chen Wang, En Huang, Jincai Wang, Fei-Yue College of Systems Engineering National University of Defense Technology Hunan Province Changsha410073 China BNRist Tsinghua University Beijing100084 China College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China

Spatial Crowdsourcing (SC) is gaining traction in both academia and industry, with tasks on SC platforms becoming increasingly complex and requiring collaboration among workers with diverse skills. Recent research works address complex tasks by dividing them into subtasks with dependencies and assigning them to suitable workers. However, the dependencies among subtasks and their heterogeneous skill requirements, as well as the need for efficient utilization of workers’ limited work time in the multi-task allocation mode, pose challenges in achieving an optimal task allocation scheme. Therefore, this paper formally investigates the problem of Dependency-aware Multi-task Allocation (DMA) and presents a well-designed framework to solve it, known as Heterogeneous Graph Reinforcement Learning-based Task Allocation (HGRL-TA). To address the challenges associated with representing and embedding diverse problem instances to ensure robust generalization, we propose a multi-relation graph model and a Compound-path-based Heterogeneous Graph Attention Network (CHANet) for effectively representing and capturing intricate relations among tasks and workers, as well as providing embedding of problem state. The task allocation decision is determined sequentially by a policy network, which undergoes simultaneous training with CHANet using the proximal policy optimization algorithm. Extensive experiment results demonstrate the effectiveness and generality of the proposed HGRL-TA in solving the DMA problem, leading to average profits that is 21.78% higher than those achieved using the metaheuristic methods. © 2024, CC BY-NC-ND.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Fast Trajectory Planning for Dubins Vehicles Under Cumulative Probability of Radar Detection

SSRN

引用

SSRN 2022年

作者： Li, Zhuo You, keyou Sun, Jian Song, Shiji State Key Lab of Intelligent Control and Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing100081 China Department of Automation and BNRist Tsinghua University Beijing100084 China Beijing Institute of Technology Chongqing Innovation Center Chongqing401120 China

This paper studies a minimum-time trajectory planning problem under radar detection, where a Dubins vehicle aims to attack a target with a limited probability of being detected. Since the probability is accumulated along the vehicle's trajectory in an integral form, we have to address a non-convex constrained functional optimization problem. To this end, Pontryagin's minimum principle is first adopted to derive the optimality condition, based on which we obtain a set of parameterized trajectories that contains all the optimal ones. Then, fast algorithms are proposed to approximately compute a minimum-time trajectory among the set. Simulations are performed to validate the effectiveness and efficiency of the proposed planning algorithms. © 2022, The Authors. All rights reserved.

关键词： Probability

来源：评论

学校读者我要写书评

暂无评论

An Earth Observation Satellite Mission Planning Method Based on Deep Q-Learning 6

An Earth Observation Satellite Mission Planning Method Based...

引用

6th International Symposium on Computer and Information Processing Technology, ISCIPT 2021

作者： Chen, Tianyuan Wang, Hongfei Liu, Hao Wu, Peng University of Chinese Academy of Sciences Key Laboratory of Space Utilization Technology and Engineering Center for Space Utilization Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences School of Artificial Intelligence Beijing China

ISBN: (纸本)9781665441377

With the increasing number of earth observation satellites and observation demands, satellite mission planning is facing huge challenges. For traditional satellite mission planning methods, there are problems such as difficulty in modeling, high computational complexity, and low scalability. This paper proposes a satellite mission planning method based on Deep Q-Learning. Firstly, a hybrid satellite mission planning model is established that combines the constraint-satisfaction model, the 0-1 knapsack model and the Markov decision model. Then, the decision algorithm based on Deep Q-Learning is used to solve the problem, so that the satellite can make full use of the past mission planning experience to learn the optimal planning strategy. Finally, several traditional satellite mission planning methods-branch and bound algorithm, genetic algorithm, simulated annealing algorithm are compared, and various performances of the algorithms are tested and analyzed through simulation experiments. The superiority of this method is verified. © 2021 IEEE.

关键词： Satellites

来源：评论

学校读者我要写书评

暂无评论

OSTER: An Orientation Sensitive Scene Text Recognizer with CenterLine Rectification 5th

OSTER: An Orientation Sensitive Scene Text Recognizer with C...

引用

5th Asian Conference on Pattern Recognition, ACPR 2019

作者： Feng, Zipeng Du, Chen Wang, Yanna Xiao, Baihua The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9783030414030

Scene texts in China are always arbitrarily arranged in two forms: horizontally and vertically. These two forms of texts exhibit distinctive features, making it difficult to recognize them simultaneously. Besides, recognizing irregular scene texts is still a challenging task due to their various shapes and distorted patterns. In this paper, we propose an orientation sensitive network aiming at distinguishing between Chinese horizontal and vertical texts. The learned orientation is then passed into an attention selective network to adjust the attention maps of the sequence recognition model, leading it working for each type of texts respectively. In addition, a lightweight centerline rectification network is adopted, which enables the irregular texts more readable while no redundant labels are needed. A synthetic dataset named SCTD is released to support our training and evaluate the proposed model. Extensive experiments show that the proposed method is capable of recognizing arbitrarily-aligned scene texts accurately and efficiently, achieving state-of-the-art performance over a number of public datasets. © Springer Nature Switzerland AG 2020.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Data-driven Self-triggered control via Trajectory Prediction

arXiv

引用

arXiv 2022年

作者： Liu, Wenjie Sun, Jian Wang, Gang Bullo, Francesco Chen, Jie The State Key Lab of Intelligent Control and Decision of Complex Systems The School of Automation Beijing Institute of Technology Beijing100081 China The Beijing Institute of Technology Chongqing Innovation Center Chongqing401120 China The Mechanical Engineering Department The Center of Control Dynamical Systems and Computation UC Santa BarbaraCA93106-5070 United States The Department of Control Science and Engineering Tongji University Shanghai201804 China The State Key Lab of Intelligent Control and Decision of Complex Systems Beijing Institute of Technology Beijing100081 China

Self-triggered control, a well-documented technique for reducing the communication overhead while ensuring desired system performance, is gaining increasing popularity. However, existing methods for self-triggered control require explicit system models that are assumed perfectly known a priori. An end-to-end control paradigm known as data-driven control learns control laws directly from data, and offers a competing alternative to the routine system identification-then-control method. In this context, the present paper puts forth data-driven self-triggered control schemes for unknown linear systems using data collected offline. Specifically, for output feedback control systems, a data-driven model predictive control (MPC) scheme is proposed, which computes a sequence of control inputs while generating a predicted system trajectory. A data-driven self-triggering law is designed using the predicted trajectory, to determine the next triggering time once a new measurement becomes available. For state feedback control systems, instead of capitalizing on MPC to predict the trajectory, a data-fitting problem using the pre-collected input-state data is solved, whose solution is employed to construct the self-triggering mechanism. Both feasibility and stability are established for the proposed self-triggered controllers, which are validated using numerical examples. © 2022, CC BY.

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

EiHi Net: Out-of-Distribution Generalization Paradigm

arXiv

引用

arXiv 2022年

作者： Wei, Qinglai Yuan, Beiming Chen, Diancheng State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China Institute of Systems Engineering Macau University of Science and Technology 999078 China

This paper develops EiHi to solve the out of distribution (O.o.D.) problem in recognition, especially the distribution shift of background. EiHi net is a model learning paradigm that can be blessed on any visual backbone. Different from the traditional deep model, namely find out correlated relationship between sample features and corresponding categories, EiHi net suffers less from spurious correlations. Fusing SimCLR and VIC-Reg via explicit and dynamical establishment of original - positive - negative sample pair as a minimal learning element, EiHi net iteratively establishes a relationship close to the causal one between features and labels, while suppresses the spurious correlations. To further reinforce the ability of the proposed model on causal reasoning, we develop a human-in-the-loop strategy, with few guidance samples, to prune the representation space directly. Finally, it is shown that the developed EiHi net makes significant improvements in the most difficult and typical O.o.D. dataset NICO, compared with the current SOTA results, without any domain (e.g. background) information. Copyright © 2022, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Learning from the Negativity: Deep Negative Correlation Meta-Learning for Adversarial Image Classification 27th

Learning from the Negativity: Deep Negative Correlation Meta...

引用

27th International Conference on MultiMedia Modeling, MMM 2021

作者： Zheng, Wenbo Yan, Lan Wang, Fei-Yue Gou, Chao School of Software Engineering Xi’an Jiaotong University Xi’an China School of Intelligent Systems Engineering Sun Yat-sen University Guangzhou China The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9783030678319

Adversarial images are commonly viewed negatively for neural network training. Here we present an opposite perspective: adversarial images can be used to investigate the problem of classifying adversarial images themselves. To this end, we propose a novel framework termed as Deep Negative Correlation Meta-Learning. In particular, we present a deep relation network to capture and memorize the relation among different samples. We further formulate a deep negative correlation learning, and design a novel meta-learning-based classifier to prevent overfitting and learn discriminative features. The final classification is derived from our relation network by learning to compare the features of samples. Experimental results demonstrate that our approach achieves significantly higher performance compared with other state-of-the-arts. © 2021, Springer Nature Switzerland AG.

关键词： Image classification

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：