检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Hu, Yan Qiu, Zhongxi Zeng, Dan Jiang, Li Lin, Chen Liu, Jiang Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Technology Southern University of Science and Technology Shenzhen518055 China China Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China

Vascular segmentation extracts blood vessels from images and serves as the basis for diagnosing various diseases, like ophthalmic diseases. Ophthalmologists often require high-resolution segmentation results for analysis, which leads to super-computational load by most existing methods. If based on low-resolution input, they easily ignore tiny vessels or cause discontinuity of segmented vessels. To solve these problems, the paper proposes an algorithm named SuperVessel, which gives out high-resolution and accurate vessel segmentation using low-resolution images as input. We first take super-resolution as our auxiliary branch to provide potential high-resolution detail features, which can be deleted in the test phase. Secondly, we propose two modules to enhance the features of the interested segmentation region, including an upsampling with feature decomposition (UFD) module and a feature interaction module (FIM) with a constraining loss to focus on the interested features. Extensive experiments on three publicly available datasets demonstrate that our proposed SuperVessel can segment more tiny vessels with higher segmentation accuracy IoU over 6%, compared with other state-of-the-art algorithms. Besides, the stability of SuperVessel is also stronger than other algorithms. We will release the code after the paper is published. Copyright © 2022, The Authors. All rights reserved.

关键词： Blood vessels

来源：评论

学校读者我要写书评

暂无评论

Dynamic and low-dimensional modeling of brain functional connectivity on Riemannian manifolds

引用

NeuroImage 2025年 314卷 121243-121243页

作者： Wang, Mingyu Wang, Yueming Yang, Yuxiao MOE Frontier Science Center for Brain Science and Brain-machine Integration Zhejiang University Hangzhou China Nanhu Brain-computer Interface Institute Hangzhou China School of Computer Science and Technology Zhejiang University Hangzhou China State Key Laboratory of Brain-machine Intelligence Zhejiang University Hangzhou China Second Affiliated Hospital School of Medicine Zhejiang University Hangzhou China

Modeling brain functional connectivity (FC) is key in investigating brain functions and dysfunctions. FC is typically quantified by symmetric positive definite (SPD) matrices that are located on a Riemannian manifold rather than the regular Euclidean space, whose modeling faces three challenges. First, FC can be time-varying and the temporal dynamics of FC matrix time-series need to be modeled within the constraint of the SPD Riemannian manifold geometry, which remains elusive. Second, the FC matrix time-series exhibits considerable stochasticity, whose probability distribution is difficult to model on the Riemannian manifold. Third, FC matrices are high-dimensional and dimensionality reduction methods for SPD matrix time-series are still lacking. Here, we develop a Riemannian state-space modeling framework to simultaneously address the challenges. First, we construct a new Riemannian state-space model (RSSM) to define a hidden SPD matrix state to achieve dynamic, stochastic, and low-dimensional modeling of FC matrix time-series on the SPD Riemannian manifold. Second, we develop a new Riemannian Particle Filter (RPF) algorithm to estimate the hidden low-dimensional SPD matrix state and predict the FC matrix time-series. Third, we develop a new Riemannian Expectation Maximization (REM) algorithm to fit the RSSM parameters. We evaluate the proposed RSSM, RPF, and REM using simulation and real-world EEG datasets, demonstrating that the RSSM enables accurate prediction of the EEG FC time-series and classification of emotional states, outperforming traditional Euclidean methods. Our results have implications for modeling brain FC on the SPD Riemannian manifold to study various brain functions and dysfunctions. © 2025 The Authors

关键词： brain functional connectivity Dimensionality reduction EEG Riemannian manifold Stochastic matrix dynamics

来源：评论

学校读者我要写书评

暂无评论

Constrained Reinforcement Learning for Dynamic Material Handling

Constrained Reinforcement Learning for Dynamic Material Hand...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Chengpeng Hu Ziming Wang Jialin Liu Junyi Wen Bifei Mao Xin Yao Research Institute of Trustworthy Autonomous Systems (RITAS) Southern University of Science and Technology Shenzhen China Department of Computer Science and Engineering Guangdong Key Laboratory of Brain-inspired Intelligent Computation Southern University of Science and Technology Shenzhen China Trustworthiness Theory Research Center Huawei Technologies Co. Ltd Shenzhen China

As one of the core parts of flexible manufacturing systems, material handling involves storage and transportation of materials between workstations with automated vehicles. The improvement in material handling can impulse the overall efficiency of the manufacturing system. However, the occurrence of dynamic events during the optimisation of task arrangements poses a challenge that requires adaptability and effectiveness. In this paper, we aim at the scheduling of automated guided vehicles for dynamic material handling. Motivated by some real-world scenarios, unknown new tasks and unexpected vehicle breakdowns are regarded as dynamic events in our problem. We formulate the problem as a constrained Markov decision process which takes into account tardiness and available vehicles as cumulative and instantaneous constraints, respectively. An adaptive constrained reinforcement learning algorithm that combines Lagrangian relaxation and invalid action masking, named RCPOM, is proposed to address the problem with two hybrid constraints. Moreover, a gym-like dynamic material handling simulator, named DMH-GYM, is developed and equipped with diverse problem instances, which can be used as benchmarks for dynamic material handling. Experimental results on the problem instances demonstrate the outstanding performance of our proposed approach compared with eight state-of-the-art constrained and non-constrained reinforcement learning algorithms, and widely used dispatching rules for material handling.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lamarckian Platform: Pushing the Boundaries of Evolutionary Reinforcement Learning towards Asynchronous Commercial Games

arXiv

引用

arXiv 2022年

作者： Bai, Hui Shen, Ruimin Lin, Yue Xu, Botian Cheng, Ran The Guangdong Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen518055 China The NetEase Games Ai Lab Guangzhou510653 China

Despite the emerging progress of integrating evolutionary computation into reinforcement learning, the absence of a high-performance platform endowing composability and massive parallelism causes non-trivial difficulties for research and applications related to asynchronous commercial games. Here we introduce Lamarckian - an open-source platform featuring support for evolutionary reinforcement learning scalable to distributed computing resources. To improve the training speed and data efficiency, Lamarckian adopts optimized communication methods and an asynchronous evolutionary reinforcement learning workflow. To meet the demand for an asynchronous interface by commercial games and various methods, Lamarckian tailors an asynchronous Markov Decision Process interface and designs an object-oriented software architecture with decoupled modules. In comparison with the state-of-the-art RLlib, we empirically demonstrate the unique advantages of Lamarckian on benchmark tests with up to 6000 CPU cores: i) both the sampling efficiency and training speed are doubled when running PPO on Google football game;ii) the training speed is 13 times faster when running PBT+PPO on Pong game. Moreover, we also present two use cases: i) how Lamarckian is applied to generating behavior-diverse game AI;ii) how Lamarckian is applied to game balancing tests for an asynchronous commercial game. Copyright © 2022, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Motor Imagery EEG Signal Classification with Simplified GoogLeNet

Enhancing Motor Imagery EEG Signal Classification with Simpl...

引用

International Winter Workshop on brain-computer interface (BCI)

作者： Lu Wang Junkongshuai Wang Bo Wen Wei Mu Lusheng Liu Jiaguan Han Lihua Zhang Jie Jia Xiaoyang Kang Shanghai Engineering Research Center of AI & Robotics MOE Frontiers Center for Brain Science State Key Laboratory of Medical NeurobiologyAcademy for Engineering & Technology Laboratory for Neural Interface and Brain Computer Interface Engineering Research Center of AI & Robotics Ministry of Education Institute of AI & Robotics Institute of Meta-MedicalFudan University Shanghai China Ji Hua Laboratory Guangdong Province Foshan China Yiwu Research Institute of Fudan University Chengbei Road Yiwu City Zhejiang China Department of Rehabilitation Medicine Huashan Hospital Fudan University Shanghai China Research Center for Intelligent Sensing Zhejiang Lab Hangzhou China

The brain-computer interface is a novel tool for interaction between the brain and external devices, and is an important tool at the forefront of international brain science research. Motor imagery does not require external stimulation and is one of the most widely used BCI paradigms, and is now widely used to address the treatment of neurological impairments such as paralysis, stroke, and acromegaly. Electrophysiological source imaging is a technique based on the observation of electromagnetic signals from the cerebral cortex, reflecting the electrophysiological activity of brain nerve cells on the scalp surface of the brain. The wavelet transform, on the other hand, is commonly used for the processing of non-smooth signals and aims to extract excellent signal features. This paper combines these two algorithms as feature extraction methods. In recent years, deep learning has been increasingly combined with MI-EEG-based BCI. GoogLeNet is one of the most classical convolutional neural networks and has made a splash in ImageNet image recognition. However, using GoogLeNet to train the BCI public dataset BCI IV IIa and dataset HGD has shown problems such as long training time, overfitting, and gradient disappearance. In response, this paper proposes an improved and streamlined version of GoogLeNet for training based on this dataset to achieve an overall improvement in accuracy, training time, number of parameters, and other indicators. It is useful for the realistic application of MI-based BCI systems.

关键词： Training Wavelet transforms Surface waves Scalp Pattern classification Electrophysiology Streaming media

来源：评论

学校读者我要写书评

暂无评论

Tagging Continuous Labels for EEG-based Emotion Classification

Tagging Continuous Labels for EEG-based Emotion Classificati...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Rong-Fei Gu Li-Ming Zhao Wei-Long Zheng Bao-Liang Lu Department of Computer Science and Engineering Center for Brain-Like Computing and Machine Intelligence the Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering and Brain Science and Technology Research Center Shanghai Jiao Tong University Shanghai People’s Republic of China RuiJin-Mihoyo Laboratory Clinical Neuroscience Center RuiJin Hospital Shanghai Jiao Tong University School of Medicine Shanghai People’s Republic of China

EEG-based emotion classification has long been a critical task in the field of affective brain-computer interface (aBCI). The majority of leading researches construct supervised learning models based on labeled datasets. Several datasets have been released, including different kinds of emotions while utilizing various forms of stimulus materials. However, they adopt discrete labeling methods, in which the EEG data collected during the same stimulus material are given a same label. These methods neglect the fact that emotion changes continuously, and mislabeled data possibly exist. The imprecision of discrete labels may hinder the progress of emotion classification in concerned works. Therefore, we develop an efficient system in this paper to support continuous labeling by giving each sample a unique label, and construct a continuously labeled EEG emotion dataset. Using our dataset with continuous labels, we demonstrate the superiority of continuous labeling in emotion classification through experiments on several classification models. We further utilize the continuous labels to identify the EEG features under induced and non-induced emotions in both our dataset and a public dataset. Our experimental results reveal the learnability and generality of the relation between the EEG features and their continuous labels.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamic network embedding via incremental skip-gram with negative sampling

引用

science China(Information sciences) 2020年第10期63卷 89-107页

作者： Hao PENG Jianxin LI Hao YAN Qiran GONG Senzhang WANG Lin LIU Lihong WANG Xiang REN Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University State Key Laboratory of Software Development Environment Beihang University Collage of Computer Science and Technology Nanjing University of Aeronautics and Astronautics National Computer Network Emergency Response Technical Team/Coordination Center of China Department of Computer Science University of Southern California

Network representation learning, as an approach to learn low dimensional representations of vertices, has attracted considerable research attention recently. It has been proven extremely useful in many machine learning tasks over large graph. Most existing methods focus on learning the structural representations of vertices in a static network, but cannot guarantee an accurate and efficient embedding in a dynamic network scenario. The fundamental problem of continuously capturing the dynamic properties in an efficient way for a dynamic network remains unsolved. To address this issue, we present an efficient incremental skip-gram algorithm with negative sampling for dynamic network embedding, and provide a set of theoretical analyses to characterize the performance guarantee. Specifically, we first partition a dynamic network into the updated, including addition/deletion of links and vertices, and the retained networks over time. Then we factorize the objective function of network embedding into the added, vanished and retained parts of the network. Next we provide a new stochastic gradient-based method, guided by the partitions of the network, to update the nodes and the parameter vectors. The proposed algorithm is proven to yield an objective function value with a bounded difference to that of the original objective function. The first order moment of the objective difference converges in order of O（1/n;）, and the second order moment of the objective difference can be stabilized in order of O（1）. Experimental results show that our proposal can significantly reduce the training time while preserving the comparable performance. We also demonstrate the correctness of the theoretical analysis and the practical usefulness of the dynamic network embedding. We perform extensive experiments on multiple real-world large network datasets over multi-label classification and link prediction tasks to evaluate the effectiveness and efficiency of the proposed framework, and up to

关键词： dynamic network embedding bound and convergence analysis multi-label classification link prediction

来源：评论

学校读者我要写书评

暂无评论

A Multi-Model Framework to Explore ADHD Diagnosis from Neuroimaging Data

引用

Journal of Data science 2024年第2期22卷 191-207页

作者： Ozdemir, Yagmur Yavuz Nukala, Naga Chandra Padmini Molinari, Roberto Deshpande, Gopikrishna Department of Mathematics and Statistics Auburn University Auburn AL United States AU MRI Research Center Department of Electrical and Computer Engineering Auburn University Auburn AL United States Department of Psychological Sciences Auburn University Auburn AL United States Alabama Advanced Imaging Consortium Auburn AL United States Center for Neuroscience Auburn University Auburn AL United States School of Psychology Capital Normal University Beijing China Key Laboratory for Learning and Cognition Capital Normal University Beijing China Department of Psychiatry National Institute of Mental Health and Neurosciences Karnataka India Centre for Brain Research Indian Institute of Science Bangalore India Department of Heritage Science and Technology Indian Institute of Technology Hyderabad India

Attention Deficit Hyperactivity Disorder (ADHD) is a frequent neurodevelopmental disorder in children that is commonly diagnosed subjectively. The objective detection of ADHD based on neuroimaging data has been a complex problem with low ranges of accuracy, possibly due to (among others) complex diagnostic processes, the high number of features considered and imperfect measurements in data collection. Hence, reliable neuroimaging biomarkers for detecting ADHD have been elusive. To address this problem we consider a recently proposed multi-model selection method called Sparse Wrapper AlGorithm (SWAG), which is a greedy algorithm that combines screening and wrapper approaches to create a set of low-dimensional models with good predictive power. While preserving the previous levels of accuracy, SWAG provides a measure of importance of brain regions for identifying ADHD. Our approach also provides a set of equally-performing and simple models which highlight the main feature combinations to be analyzed and the interactions between them. Taking advantage of the network of models resulting from this approach, we confirm the relevance of the frontal and temporal lobes as well as highlight how the different regions interact to detect the presence of ADHD. In particular, these results are fairly consistent across different learning mechanisms employed within the SWAG (i.e. logistic regression, linear and radial-kernel support vector machines) thereby providing population-level insights, as well as delivering feature combinations that are smaller and often perform better than those that would be used if employing their original versions directly. © 2024 The Author(s).

关键词： automated detection functional magnetic resonance imaging interpretability prediction accuracy SWAG

来源：评论

学校读者我要写书评

暂无评论

Electromyography-Based Intentional-Deception Behavior Analysis in an Interactive Social Context: Statistical Analysis and Machine Learning

SSRN

引用

SSRN 2023年

作者： Dong, Zizhao Li, Jingting Yang, Xingpeng Guo, Changfu Lu, Shaoyuan Wang, Su-Jing Fu, Xiaolan CAS Key Laboratory of Behavioral Science Institute of Psychology Beijing100101 China Department of Psychology University of the Chinese Academy of Sciences Beijing101408 China School of Computer Science Jiangsu University of Science and Technology Zhenjiang212003 China School of Psychology Capital Normal University Beijing100048 China State Key Laboratory of Brain and Cognitive Science Institute of Psychology CAS Beijing100101 China

Lying is a common social behavior, and accurate lie detection is crucial in areas such as national security. However, existing lie detection techniques have certain limitations. Therefore, more accurate and reliable tools and methods are needed to meet the practical needs of lie detection. In this context, this study discovered the potential value of electromyography (EMG) as a lie detection indicator. Specifically, this study used EMG for statistical analysis and machine learning recognition analysis of the lying process in an interactive scenario of active lying. Furthermore, we compared the performance of two traditional machine learning models and one deep learning model for lie detection based on EMG signals. In particular, time-dimensional and time-frequency-dimensional EMG features were used to mine and lie related features. Statistical results showed that compared to truth-telling, people tend to suppress their facial expressions when preparing to lie. Some facial muscle movements that were not be successfully suppressed after lying may be crucial for detecting lies. Moreover, our study offers theoretical hypotheses for the occurrence of micro-expressions and the feature of upper-lower facial asymmetry. Besides the statistic analysis, the analysis results of machine learning also demonstrated demonstrate the potential of machine learning models for EMG-based intelligent lying process analysis, particularly the RUSBoosted tree. In addition, our experiment result also proved that focusing on specific facial muscles, such as Corrugator supercilii, could improve the accuracy and efficiency of intelligent algorithms. In summary, our research results provide more insights into the cognitive and facial muscle movement patterns involved in lying based on statistical analysis and machine learning. © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Provably Convergent Subgraph-wise Sampling for Fast GNN Training

arXiv

引用

arXiv 2023年

作者： Wang, Jie Shi, Zhihao Liang, Xize Lian, Defu Ji, Shuiwang Li, Bin Chen, Enhong Wu, Feng MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition University of Science and Technology of China Hefei230027 China CAS Key Laboratory of Technology in GIPAS University of Science and Technology of China Hefei230027 China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230091 China University of Science and Technology of China Hefei230027 China The Department of Computer Science and Engineering Texas A&M University College StationTX77843 United States

Subgraph-wise sampling—a promising class of mini-batch training techniques for graph neural networks (GNNs)—is critical for real-world applications. During the message passing (MP) in GNNs, subgraph-wise sampling methods discard messages outside the mini-batches in backward passes to avoid the well-known neighbor explosion problem, i.e., the exponentially increasing dependencies of nodes with the number of MP iterations. However, discarding messages may sacrifice the gradient estimation accuracy, posing significant challenges to their convergence analysis and convergence speeds. To address this challenge, we propose a novel subgraph-wise sampling method with a convergence guarantee, namely Local Message Compensation (LMC). To the best of our knowledge, LMC is the first subgraph-wise sampling method with provable convergence. The key idea is to retrieve the discarded messages in backward passes based on a message passing formulation of backward passes. By efficient and effective compensations for the discarded messages in both forward and backward passes, LMC computes accurate mini-batch gradients and thus accelerates convergence. Moreover, LMC is applicable to various MP-based GNN architectures, including convolutional GNNs (finite message passing iterations with different layers) and recurrent GNNs (infinite message passing iterations with a shared layer). Experiments on large-scale benchmarks demonstrate that LMC is significantly faster than state-of-the-art subgraph-wise sampling methods. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

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：