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

学校读者我要写书评

暂无评论

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

Hyperbolic variational graph neural network for modeling dynamic graphs

学校读者我要写书评

暂无评论

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

Signal approximation with Pascal’s triangle and sampling

学校读者我要写书评

暂无评论

Signal approximation with Pascal’s triangle and sampling

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

Graph classification based on skeleton and component features

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Xue Wei, Wei Feng, Xiangnan Cao, Xiaobo Sun, Dan Beijing System Design Institute of Electro-Mechanic Engineering Beijing100854 China School of Mathematical Sciences Beihang University Beijing100191 China Key Laboratory of Mathematics Informatics and Behavioral Semantics Ministry of Education 100191 China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China Peng Cheng Laboratory Shenzhen Guangdong518066 China

Most existing popular methods for learning graph embedding only consider fixed-order global structural features and lack structures hierarchical representation. To address this weakness, we propose a novel graph embedding algorithm named GraphCSC that realizes classification based on skeleton information using fixed-order structures learned in anonymous random walks manner, and component information using different size subgraphs. Two graphs are similar if their skeletons and components are both similar, thus in our model, we integrate both of them together into embeddings as graph homogeneity characterization. We demonstrate our model on different datasets in comparison with a comprehensive list of up-to-date state-of-the-art baselines, and experiments show that our work is superior in real-world graph classification tasks. © 2021, CC BY.

关键词： Musculoskeletal system

An Anomaly Pattern Detection Method for Sensor data 16th

学校读者我要写书评

暂无评论

An Anomaly Pattern Detection Method for Sensor Data

16th Web Information Systems and Applications Conference, WISA 2019

作者： Li, Han Yu, Bin Zhao, Ting College of Computer Science North China University of Technology Beijing China Beijing Key Laboratory on Integration and Analysis of Large-Scale Stream Data Beijing China Advanced Computing and Big Data Technology Laboratory of SGCC Global Energy Interconnection Research Institute Beijing China

ISBN: (纸本)9783030309510

With the development of the Internet of Things (IOT) technology, a large number of sensor data have been produced. Due to the complex acquisition environment and transmission condition, anomalies are prevalent. Sensor data is a kind of typical time series data, its anomaly refers to not only outliers, but also the anomaly of continuous data fragments, namely anomaly patterns. To achieve anomaly pattern detection on sensor data, the characteristics of sensor data are analyzed including temporal correlation, spatial correlation and high dimension. Then based on these characteristics and the real-time processing requirements of sensor data, a sensor data oriented anomaly pattern detection approach is proposed in this paper. In the approach, the frequency domain features of sensor data are obtained by Fast Fourier Transform, the dimension of the feature space is reduced by describing frequency domain features with statistical values, and the high-dimensional sensor data is processed in time on the basis of Isolation Forest algorithm. In order to verify the feasibility and effectiveness of the proposed approach, experiments are carried out on the open dataset IBRL. The experimental results show that the approach can effectively identify the pattern anomalies of sensor data, and has low time cost while ensuring the high accuracy. © 2019, Springer Nature Switzerland AG.

关键词： Fast Fourier transforms

Manipulating density of magnetic skyrmions via multilayer repetition and thermal annealing

学校读者我要写书评

暂无评论

Physical Review B 2021年第6期104卷 064421-064421页

作者： Xinran Wang Anni Cao Sai Li Jin Tang Ao Du Houyi Cheng Yiming Sun Haifeng Du Xueying Zhang Weisheng Zhao Fert Beijing Institute School of Integrated Circuit Science and Engineering Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing 100191 China Beihang-Goertek Joint Microelectronics Institute Qingdao Research Institute Beihang University Qingdao 266000 China Beijing Microelectronics Technology Institute Beijing 100076 China Anhui Province Key Laboratory of Condensed Matter Physics at Extreme Conditions High Magnetic Field Laboratory of Chinese Academy of Sciences and University of Science and Technology of China Hefei 230031 China

The magnetic skyrmion, a tiny magnetic texture that holds promise as the next-generation information carrier, has been widely studied in recent years. A fine tunability of skyrmion density is required for its real applications in novel memory and logic devices. Here, we report on the manipulation of skyrmion density at room temperature in a Pt/Co/Ta/MgO system composed of multiple repetitions with perpendicular magnetic anisotropy. Néel-type skyrmions are observed by performing Lorentz transmission electron microscopy and magnetic force microscopy. The influence of structure repetition on skyrmions is experimentally investigated and further understood by micromagnetic simulations. The variation of skyrmion density with repetition number is mainly attributed to magnetic anisotropy. Meanwhile, thermal annealing is utilized to regulate the skyrmion density, where skyrmion-related properties exert a competitive effect. Our findings provide alternative means to manipulate skyrmion density, further allowing for optimized engineering of skyrmion-based devices.

关键词： Magnetic interactions Magnetic texture Skyrmions Spintronics Multilayer thin films Brillouin scattering & spectroscopy Lorentz microscopy Magnetic force microscopy

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

学校读者我要写书评

暂无评论

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

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

Representation Learning of Graphs Using Graph Convolutional Multilayer Networks Based on Motifs

学校读者我要写书评

暂无评论

arXiv 2020年

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

The graph structure is a commonly used data storage mode, and it turns out that the low-dimensional embedded representation of nodes in the graph is extremely useful in various typical tasks, such as node classification, link prediction, etc. However, most of the existing approaches start from the binary relationship (i.e., edges) in the graph and have not leveraged the higher order local structure (i.e., motifs) of the graph. Here, we propose mGCMN – a novel framework which utilizes node feature information and the higher order local structure of the graph to effectively generate node embeddings for previously unseen data. Through research we have found that different types of networks have different key motifs. And the advantages of our method over the baseline methods have been demonstrated in a large number of experiments on citation network and social network datasets. At the same time, a positive correlation between increase of the classification accuracy and the clustering coefficient is revealed. It is believed that using high order structural information can truly manifest the potential of the network, which will greatly improve the learning efficiency of the graph neural network and promote a brand-new learning mode establishment. Copyright © 2020, The Authors. All rights reserved.

关键词： Graph neural networks

Cross-domain object detection through coarse-to-fine feature adaptation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zheng, Yangtao Huang, Di Liu, Songtao Wang, Yunhong 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 Beijing100191 China

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 coarseto-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. Copyright © 2020, The Authors. All rights reserved.

关键词： Object recognition