检索结果-内蒙古大学图书馆

Understanding collective behaviors in reinforcement learning evolutionary games via a belief-based formalization

Physical Review E 2020年第4期101卷 042402-042402页

作者： Ji-Qiang Zhang Si-Ping Zhang Li Chen Xu-Dong Liu Beijing Advanced Innovation Center for Big Data and Brain Computing School of Comuter Science and Engineering Beihang University Beijing 100191 China The Key Laboratory of Biomedical Information Engineering of Ministry of Education The Key Laboratory of Neuro-informatics & Rehabilitation Engineering of Ministry of Civil Affairs and Institute of Health and Rehabilitation Science School of Life Science and Technology Xi'an Jiaotong University Xi'an 710049 China School of Physics and Information Technology Shaanxi Normal University Xi'an 710062 China Beijing Advanced Innovation Center for Big Data and Brain Computing School of Computer Science and Engineering Beihang University Beijing 100191 China

Collective behaviors by self-organization are ubiquitous in nature and human society and extensive efforts have been made to explore the mechanisms behind them. Artificial intelligence (AI) as a rapidly developing field is of great potential for these tasks. By combining reinforcement learning with evolutionary game (RLEG), we numerically discover a rich spectrum of collective behaviors—explosive events, oscillation, and stable states, etc., that are also often observed in the human society. In this work, we aim to provide a theoretical framework to investigate the RLEGs systematically. Specifically, we formalize AI-agents' learning processes in terms of belief switches and behavior modes defined as a series of actions following beliefs. Based on the preliminary results in the time-independent environment, we investigate the stability at the mixed equilibrium points in RLEGs generally, in which agents reside in one of the optimal behavior modes. Moreover, we adopt the maximum entropy principle to infer the composition of agents residing in each mode at a strictly stable point. When the theoretical analysis is applied to the 2×2 game setting, we can explain the uncovered collective behaviors and are able to construct equivalent systems intuitively. Also, the inferred composition of different modes is consistent with simulations. Our work may be helpful to understand the related collective emergence in human society as well as behavioral patterns at the individual level and potentially facilitate human-computer interactions in the future.

关键词： Collective dynamics Complex systems Evolutionary dynamics

来源：评论

学校读者我要写书评

暂无评论

Adaptive NMS: Refining Pedestrian Detection in a Crowd

Adaptive NMS: Refining Pedestrian Detection in a Crowd

引用

IEEE/CVF Conference on Computer Vision and Pattern Recognition

作者： Songtao Liu Di Huang Yunhong Wang Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University

ISBN: (纸本)9781728132945

Pedestrian detection in a crowd is a very challenging issue. This paper addresses this problem by a novel Non-Maximum Suppression (NMS) algorithm to better refine the bounding boxes given by detectors. The contributions are threefold: (1) we propose adaptive-NMS, which applies a dynamic suppression threshold to an instance, according to the target density;(2) we design an efficient subnetwork to learn density scores, which can be conveniently embedded into both the single-stage and two-stage detectors;and (3) we achieve state of the art results on the CityPersons and CrowdHuman benchmarks.

关键词： state of the art Crowding Pedestrians bounding boxes Purification detector Refining Subnetwork Self tuning

来源：评论

学校读者我要写书评

暂无评论

Dynamics of all-optically switched magnetic domains in Co/Gd heterostructures with Dzyaloshinskii-Moriya interaction

引用

Physical Review B 2020年第10期102卷 104412-104412页

作者： Anni Cao Youri L. W. van Hees Reinoud Lavrijsen Weisheng Zhao Bert Koopmans Department of Applied Physics Institute for Photonic Integration Eindhoven University of Technology PO Box 513 5600 MB Eindhoven The Netherlands Fert Beijing Institute Beijing Advanced Innovation Center for Big Data and Brain Computing School of Microelectronics Beihang University Beijing 100191 China

Given the development of hybrid spintronic-photonic devices and optical manipulation of chiral magnetic textures, a combined interest in single-pulse all-optical switching (AOS) of magnetization and current-induced domain wall motion in synthetic ferrimagnetic structures with strong Dzyaloshinskii-Moriya Interaction (DMI) is emerging. In this work, we explore the role of the DMI on the AOS process as well as the stability of optically written micromagnetic domains using specially engineered Co/Gd-based multilayer structures. Quantitative insight is obtained by measuring the interesting dynamics in moon-shaped structures written by two successive laser pulses. The stability of domains resulting from an interplay between the dipolar interaction and domain-wall energy are compared to simple analytical models and micromagnetic simulations. A shortening process occurring at ns time scale is confirmed by our computational results, and a stabilizing role of DMI on microscopic AOS-written domains is experimentally demonstrated.

关键词： Light-induced magnetic effects Magnetism Single- and few-photon ionization & excitation Spintronics Ultrafast phenomena

来源：评论

学校读者我要写书评

暂无评论

Representation learning of reconstructed graphs using random walk graph convolutional network

arXiv

引用

arXiv 2021年

作者： Li, Xing Wei, Wei Feng, Xiangnan Zheng, Zhiming School of Mathematical Science Beihang University Beijing China Key Laboratory of Mathematics Informatics Behavioral Semantics Ministry of Education China Peng Cheng Laboratory ShenzhenGuangdong China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China Key Laboratory of Mathematics Informatics Behavioral Semantics Ministry of Education China

Graphs are often used to organize data because of their simple topological structure, and therefore play a key role in machine learning. And it turns out that the low-dimensional embedded representation obtained by graph representation learning are extremely useful in various typical tasks, such as node classification, content recommendation and link prediction. However, the existing methods mostly start from the microstructure (i.e., the edges) in the graph, ignoring the mesoscopic structure (high-order local structure). Here, we propose wGCN – a novel framework that utilizes random walk to obtain the node-specific mesoscopic structures of the graph, and utilizes these mesoscopic structures to reconstruct the graph And organize the characteristic information of the nodes. Our method can effectively generate node embeddings for previously unseen data, which has been proven in a series of experiments conducted on citation networks and social networks (our method has advantages over baseline methods). We believe that combining high-order local structural information can more efficiently explore the potential of the network, which will greatly improve the learning efficiency of graph neural network and promote the establishment of new learning models. © 2021, CC BY.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Hyperbolic variational graph neural network for modeling dynamic graphs

arXiv

引用

arXiv 2021年

作者： Sun, Li Zhang, Zhongbao Zhang, Jiawei Wang, Feiyang Peng, Hao Su, Sen Yu, Philip S. State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China IFM Lab Department of Computer Science Florida State University FL United States Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University China Department of Computer Science University of Illinois ChicagoIL United States

Learning representations for graphs plays a critical role in a wide spectrum of downstream applications. In this paper, we summarize the limitations of the prior works in three folds: representation space, modeling dynamics and modeling uncertainty. To bridge this gap, we propose to learn dynamic graph representation in hyperbolic space, for the first time, which aims to infer stochastic node representations. Working with hyperbolic space, we present a novel Hyperbolic Variational Graph Neural Network, referred to as HVGNN. In particular, to model the dynamics, we introduce a Temporal GNN (TGNN) based on a theoretically grounded time encoding approach. To model the uncertainty, we devise a hyperbolic graph variational autoencoder built upon the proposed TGNN to generate stochastic node representations of hyperbolic normal distributions. Furthermore, we introduce a reparameterisable sampling algorithm for the hyperbolic normal distribution to enable the gradient-based learning of HVGNN. Extensive experiments show that HVGNN outperforms state-of-the-art baselines on real-world datasets. © 2021, CC BY-NC-SA.

关键词： Normal distribution

来源：评论

学校读者我要写书评

暂无评论

Signal approximation with Pascal’s triangle and sampling

Signal approximation with Pascal’s triangle and sampling

引用

第32届中国控制与决策会议

作者： Lei Chen Xinghuo Yu Jinhu Lü School of Automation Science and Electrical Engineering State Key Laboratory of Software Development EnvironmentBeijing Advanced Innovation Center for Big Data and Brain ComputingBeihang University School of Engineering RMIT University

This brief explores the approximation properties of a unique basis expansion based on Pascal’s triangle,which realizes a sampled-data driven approach between a continuous-time signal and its discrete-time *** roles of certain parameters,such as sampling time interval or model order,and signal characteristics,i.e.,its curvature,on the approximation are *** errors in one and multiple-step predictions are ***,time-variant approximations under the thresholds of signal curvature are employed to narrow errors and provide flexibilities.

关键词： Signal approximation Sampling Discretization

来源：评论

学校读者我要写书评

暂无评论

Theoretical conditions for field-free magnetization switching induced by spin-orbit torque and dzyaloshinskii-moriya interaction

arXiv

引用

arXiv 2020年

作者： Wang, Min Wang, Zhaohao Zhang, Xueying Zhao, Weisheng Fert Beijing Research Institute Beihang University Beijing China School of Microelectronics Beihang University Beijing China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China

Recently, it was demonstrated that field-free switching could be achieved by combining spin-orbit torque (SOT) and Dzyaloshinskii-Moriya interaction (DMI). However, this mechanism only occurs under certain conditions which have not been well discussed. In this letter, it is found that the ratio of domain wall width to diameter of nanodots could be utilized as a criteria for judging this mechanism. Influences of different magnetic parameters are studied, including exchange constant, DMI magnitude and field-like toque, etc. Besides, we reveal the importance of the shrinkage of magnetic domain wall surface energy for metastable states. Our work provides guidelines to the experiments on the DMI-induced field-free magnetization switching, and also offers a new way for the design of SOT-based memory or logic circuits. Copyright © 2020, The Authors. All rights reserved.

关键词： Torque

来源：评论

学校读者我要写书评

暂无评论

Event extraction by associating event types and argument roles

arXiv

引用

arXiv 2021年

作者： Li, Qian Guo, Shu Wu, Jia Li, Jianxin Sheng, Jiawei Wang, Lihong Dong, Xiaohan Peng, Hao Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China National Computer Network Emergency Response Technical Team Coordination Center of China Beijing China Department of Computing Macquarie University Sydney Australia Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China

Event extraction (EE), which acquires structural event knowledge from texts, can be divided into two sub-tasks: event type classification and element extraction (namely identifying triggers and arguments under different role patterns). As different event types always own distinct extraction schemas (i.e., role patterns), previous work on EE usually follows an isolated learning paradigm, performing element extraction independently for different event types. It ignores meaningful associations among event types and argument roles, leading to relatively poor performance for less frequent types/roles. This paper proposes a novel neural association framework for the EE task. Given a document, it first performs type classification via constructing a document-level graph to associate sentence nodes of different types, and adopting a graph attention network to learn sentence embeddings. Then, element extraction is achieved by building a universal schema of argument roles, with a parameter inheritance mechanism to enhance role preference for extracted elements. As such, our model takes into account type and role associations during EE, enabling implicit information sharing among them. Experimental results show that our approach consistently outperforms most state-of-the-art EE methods in both sub-tasks. Particularly, for types/roles with less training data, the performance is superior to the existing methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Adaptive Leaderless Consensus for a Class of Uncertain Nonlinear Systems with Intermittent Actuator Faults*

Adaptive Leaderless Consensus for a Class of Uncertain Nonli...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Marwah Saad Mahdi AL-Adhami Jiang Long Wei Wang Lei Wang School of Automation Science and Electrical Engineering Beihang University Beijing China School of Automation Science and Electrical Engineering Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China

ISBN: (数字)9781728151694

ISBN: (纸本)9781728151700

This paper investigates the leaderless consensus problem for high-order nonlinear systems with unmatched unknown parameters and intermittent actuator faults. Under directed topology condition, a novel adaptive fault tolerant leaderless consensus control scheme is proposed. In order to achieve the leaderless consensus under directed topology condition, an auxiliary system, which is related to the individual states and neighboring states, is introduced in each agent. On the other hand, to compensate the effect of intermittent actuator faults on the system stability of each agent, a local parameter estimator is introduced to estimate a lumped unknown parameter related to the lower bound of the effectiveness of actuator. It is shown that with the proposed adaptive fault tolerant leaderless consensus control scheme, all the closed-loop signals are globally uniformly bounded and all the agents' outputs can reach a consensus.

关键词： Industrial electronics Actuators Fault tolerance Fault tolerant systems Topology Nonlinear systems Numerical stability

来源：评论

学校读者我要写书评

暂无评论

Cross-domain Object Detection through Coarse-to-Fine Feature Adaptation

Cross-domain Object Detection through Coarse-to-Fine Feature...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Yangtao Zheng Di Huang Songtao Liu Yunhong Wang Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University State Key Laboratory of Software Development Environment Beihang University School of Computer Science and Engineering Beihang University Beijing China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Recent years have witnessed great progress in deep learning based object detection. However, due to the domain shift problem, applying off-the-shelf detectors to an unseen domain leads to significant performance drop. To address such an issue, this paper proposes a novel coarse-to-fine feature adaptation approach to cross-domain object detection. At the coarse-grained stage, different from the rough image-level or instance-level feature alignment used in the literature, foreground regions are extracted by adopting the attention mechanism, and aligned according to their marginal distributions via multi-layer adversarial learning in the common feature space. At the fine-grained stage, we conduct conditional distribution alignment of foregrounds by minimizing the distance of global prototypes with the same category but from different domains. Thanks to this coarse-to-fine feature adaptation, domain knowledge in foreground regions can be effectively transferred. Extensive experiments are carried out in various cross-domain detection scenarios. The results are state-of-the-art, which demonstrate the broad applicability and effectiveness of the proposed approach.

关键词： Feature extraction Subspace constraints Object detection Detectors Task analysis Semantics Prototypes

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：