检索结果-南通市图书馆

GPS: graph contrastive learning via multi-scale augmented views from adversarial pooling

Science China(information Sciences) 2025年第1期68卷 145-158页

作者： Wei JU Yiyang GU Zhengyang MAO Ziyue QIAO Yifang QIN Xiao LUO Hui XIONG Ming ZHANG School of Computer Science National Key Laboratory for Multimedia Information ProcessingPeking University Artificial Intelligence Thrust The Hong Kong University of Science and Technology Department of Computer Science University of California

Self-supervised graph representation learning has recently shown considerable promise in a range of fields, including bioinformatics and social networks. A large number of graph contrastive learning approaches have shown promising performance for representation learning on graphs, which train models by maximizing agreement between original graphs and their augmented views(i.e., positive views). Unfortunately, these methods usually involve pre-defined augmentation strategies based on the knowledge of human experts. Moreover, these strategies may fail to generate challenging positive views to provide sufficient supervision signals. In this paper, we present a novel approach named graph pooling contrast(GPS) to address these *** by the fact that graph pooling can adaptively coarsen the graph with the removal of redundancy, we rethink graph pooling and leverage it to automatically generate multi-scale positive views with varying emphasis on providing challenging positives and preserving semantics, i.e., strongly-augmented view and weakly-augmented view. Then, we incorporate both views into a joint contrastive learning framework with similarity learning and consistency learning, where our pooling module is adversarially trained with respect to the encoder for adversarial robustness. Experiments on twelve datasets on both graph classification and transfer learning tasks verify the superiority of the proposed method over its counterparts.

关键词： graph representation learning graph neural networks graph contrastive learning graph augmentations graph pooling

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Energy injection,transport,and dissipation in Earth's magnetosphere

引用

Science China Earth Sciences 2025年

作者： Huishan FU Jinbin CAO School of Space and Earth Sciences Beihang University Key laboratory of Space Environment Monitoring and information processing Ministry of Industry and Information Technology

The Earth's magnetosphere is a region occupied by many artificial satellites, and also is a region that spacecraft must cruise during the deep-space exploration. In this sense, the Earth's magnetosphere is closely related to human activity and is a candidate for us to expand our living space. Generally, the Earth's magnetosphere can preclude most energetic particles from the Sun and the interstellar space, effectively protecting human beings on the Earth from being attacked and thus making the Earth to be a habitable planet. However, in some conditions, the Earth's magnetosphere becomes dynamic and energetic, and consequently may damage the artificial satellites, threaten the astronauts' health, and disrupt the ground infrastructure, which leads to a decline in the national economy. Therefore, investigating how energy is injected into the magnetosphere, how it is transported in the magnetosphere, and how it is ultimately dissipated in the magnetosphere are the key issues in space physics. Targeting these key issues, in this paper, we review the recent progress on them. Particularly, we introduce the relevant scientific questions,models, methods, and spacecraft missions, for better building a physical link among the energy injection, transport, and dissipation in the magnetosphere, present an energy chain of the magnetosphere, reveal the relationship between such energy chain and the space weather events, and discuss the forecasting and warning methods for energetic-particle events in the magnetosphere. The magnetospheric energy chain discussed in this paper will help us reveal the mechanisms of space weather events, establish the models of space environment, and forecast the disastrous space weather events.

关键词：

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Decentralized fault diagnosis of discrete-event systems with unreliable sensors using linear temporal logic

引用

Science China(information Sciences) 2025年第1期68卷 322-341页

作者： Weijie DONG Shaoyuan LI Xiang YIN Key laboratory of System Control and information processing Ministry of EducationDepartment of Automation Shanghai Jiao Tong University

In this paper,we investigate a decentralized diagnosis problem of a discrete-evnt system(DES) subject to unreliable sensors,where the sensor observations of local diagnosers may be non-deterministic as a result of possible *** studies on decentralized robust diagnosis can only deal with different types of sensor failures separately,e.g.,all sensors suffer from the same type of sensor failures such as intermittent sensor failures or permanent sensor ***,since sensors of different local diagnosers may face different external environments and have different internal characteristics,sensors corresponding to different local diagnosers may also have their own fault *** this paper,we propose a flexible framework of decentralized diagnosis for DES subject to unreliable sensors such that sensors of different local diagnosers are permitted to have different types of sensor *** this end,we extend the existing decentralized diagnosis framework to the case where there exist common sensors broadcasting their observations to all local *** apply linear temporal logic(LTL) to constrain infinite behaviors of private sensors of local diagnosers and common ***,a new notion of φ-codiagnosability is proposed as the necessary and sufficient condition for the existence of a decentralized diagnoser that works correctly under sensors,satisfying LTL-based sensor ***,we provide an effective approach to verify the φ-codiagnosability.

关键词： discrete event systems decentralized diagnosis codiagnosability automata

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

3D facial motion tracking by combining online appearance model and cylinder head model in particle filtering

引用

Science China(information Sciences) 2014年第2期57卷 278-284页

作者： YU Jun WANG ZengFu national laboratory of speech and language information processing

A 3D facial motion tracking approach has been proposed based on the incorporation of online appearance model(OAM)and cylinder head model(CHM)in the framework of particle filtering.1)For the construction of OAM,multi-measurements are infused to reduce the influence of lighting and person dependence.2)The global motion acquired from CHM fitting is set as the initialization of OAM fitting,and the fitting result is set as the initialization of CHM in the next frame.3)Motion filtering is applied with particle filter combined with local optimization and improved *** between subjects evaluations show that our approach is fit to track facial motions.

关键词： facial motion tracking online appearance model cylinder head model particle filtering information fusion

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

DKP-SLAM:A Visual SLAM for Dynamic Indoor Scenes Based on Object Detection and Region Probability

引用

Computers, Materials & Continua 2025年第1期82卷 1329-1347页

作者： Menglin Yin Yong Qin Jiansheng Peng College of Automation Guangxi University of Science and TechnologyLiuzhou545000China Department of Artificial Intelligence and Manufacturing Hechi UniversityHechi546300China Key laboratory of AI and information processing Education Department of Guangxi Zhuang Autonomous RegionHechi546300China Guangxi Key laboratory of Sericulture Ecology and Applied Intelligent Technology School of Chemistry and BioengineeringHechi UniversityHechi546300China

In dynamic scenarios,visual simultaneous localization and mapping(SLAM)algorithms often incorrectly incorporate dynamic points during camera pose computation,leading to reduced accuracy and *** paper presents a dynamic SLAM algorithm that leverages object detection and regional dynamic ***,a parallel thread employs the YOLOX object detectionmodel to gather 2D semantic information and compensate for missed ***,an improved K-means++clustering algorithm clusters bounding box regions,adaptively determining the threshold for extracting dynamic object contours as dynamic points *** process divides the image into low dynamic,suspicious dynamic,and high dynamic *** the tracking thread,the dynamic point removal module assigns dynamic probability weights to the feature points in these *** with geometric methods,it detects and removes the dynamic *** final evaluation on the public TUM RGB-D dataset shows that the proposed dynamic SLAM algorithm surpasses most existing SLAM algorithms,providing better pose estimation accuracy and robustness in dynamic environments.

关键词： Visual SLAM dynamic scene YOLOX K-means++clustering dynamic probability

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Handling OOV Words in Mandarin Spoken Term Detection with an Hierarchical n-Gram language Model

引用

Chinese Journal of Electronics 2017年第6期26卷 1239-1244页

作者： WANG Xuyang ZHANG Pengyuan NA Xingyu PAN Jielin YAN Yonghong The Key laboratory of speech Acoustics and Content Understanding Chinese Academy of Sciences Xinjiang laboratory of Minority speech and language information processing Chinese Academy of Sciences

In this paper, an hierarchical n-gram language model(LM) combining words and characters is explored to improve the detection of Out-of-vocabulary(OOV) words in Mandarin Spoken term detection(STD).The hierarchical LM is based on a word-level LM, with a character-level LM estimating probabilities of OOV words in a class-based way. The region containing OOV words in the sentence to be decoded is detected with the help of the word-level LM and the probabilities of OOV words are derived from the character-level LM. The implementation of the proposed approach is based on a dynamic decoder. The proposed approach is evaluated in terms of Actual term weighted value(ATWV) on two Mandarin data sets. Experiment results show that more than 10% relative improvement for OOV word detection is achieved on both sets. In addition, the detection of In-vocabulary(IV) words is barely influenced as well.

关键词： Spoken term detection(STD) language model(LM) Out-of-vocabulary(OOV) words

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Single-Channel speech Separation Based on Non-negative Matrix Factorization and Factorial Conditional Random Field

引用

Chinese Journal of Electronics 2018年第5期27卷 1063-1070页

作者： LI Xu TU Ming WANG Xiaofei WU Chao FU Qiang YAN Yonghong Key laboratory of speech Acoustics and Content Understanding Institute of AcousticsChinese Academy of Sciences Signal Analysis Representation and Perception laboratory Arizona State University Xinjiang laboratory of Minority speech and language information processing

A new Non-negative matrix factorization(NMF) based algorithm is proposed for single-channel speech separation with a prior known speakers, which aims to better model the spectral structure and temporal continuity of speech signal. First, NMF and k-means clustering are employed to obtain multiple small dictionaries as well as a state sequence that describes the temporal dynamics between these dictionaries for each ***, a Factorial conditional random field(FCRF) model is trained using the state sequences and dictionaries to jointly model the temporal continuity of two speakers’ mixed signal for separation. Experiments show that the proposed algorithm outperforms the baselines with respect to all metrics, for example sparse NMF(+1.12 dB SDR, +2.37 dB SIR, +0.40 dB SAR, +0.2 MOS), nonnegative factorial hidden Markov model(+2.04 dB SDR,+4.26 dB SIR, +0.62 dB SAR, +1.0 MOS) and standard NMF(+2.8 dB SDR, +5.08 dB SIR, +1.06 dB SAR, +1.2 MOS).

关键词： Single-channel speech separation Non-negative matrix factorization Factorial conditional random field k-means clustering

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Optimizing Fine-Tuning in Quantized language Models:An In-Depth Analysis of Key Variables

引用

Computers, Materials & Continua 2025年第1期82卷 307-325页

作者： Ao Shen Zhiquan Lai Dongsheng Li Xiaoyu Hu national Key laboratory of Parallel and Distributed Computing National University of Defense TechnologyChangsha410073China Strategic Assessments and Consultation Institute Academy of Military ScienceBeijing100091China

Large-scale language Models(LLMs)have achieved significant breakthroughs in Natural language processing(NLP),driven by the pre-training and fine-tuning *** this approach allows models to specialize in specific tasks with reduced training costs,the substantial memory requirements during fine-tuning present a barrier to broader ***-Efficient Fine-Tuning(PEFT)techniques,such as Low-Rank Adaptation(LoRA),and parameter quantization methods have emerged as solutions to address these challenges by optimizing memory usage and computational *** these,QLoRA,which combines PEFT and quantization,has demonstrated notable success in reducing memory footprints during fine-tuning,prompting the development of various QLoRA *** these advancements,the quantitative impact of key variables on the fine-tuning performance of quantized LLMs remains *** study presents a comprehensive analysis of these key variables,focusing on their influence across different layer types and depths within LLM *** investigation uncovers several critical findings:(1)Larger layers,such as MLP layers,can maintain performance despite reductions in adapter rank,while smaller layers,like self-attention layers,aremore sensitive to such changes;(2)The effectiveness of balancing factors depends more on specific values rather than layer type or depth;(3)In quantization-aware fine-tuning,larger layers can effectively utilize smaller adapters,whereas smaller layers struggle to do *** insights suggest that layer type is a more significant determinant of fine-tuning success than layer depth when optimizing quantized ***,for the same discount of trainable parameters,reducing the trainable parameters in a larger layer is more effective in preserving fine-tuning accuracy than in a smaller *** study provides valuable guidance for more efficient fine-tuning strategies and opens avenues for further research into optimizing LLM

关键词： Large-scale language Model Parameter-Efficient Fine-Tuning parameter quantization key variable trainable parameters experimental analysis

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

A Stochastic Approximation Method with Enhanced Robustness for Crosstalk Cancellation

引用

Chinese Journal of Electronics 2017年第6期26卷 1269-1275页

作者： XU Huaxing WANG Qia XIA Risheng LI Junfeng YAN Yonghong Key laboratory of speech Acoustics and Content Understanding Institute of Acoustics Chinese Academy of Sciences China Long Yuan Power Group Corporation Limited Xinjiang laboratory of Minority speech and language information processing

The objective of acoustic crosstalk cancellation is to use loudspeakers to deliver prescribed binaural signals(that reproduce a particular auditory scene) to a listener’s ears, which is useful for 3-D audio *** practice, the actual transfer function matrix will differ from the design matrix, because of either the listener’s head movement or rotation, etc. Crosstalk cancellation system(CCS) is very non-robust to these perturbations. Generally, in order to improve the robustness of CCS, several pairs of loudspeakers are needed whose position varies continuously as frequency varies. With the help of assumed stochastic analysis, we propose a stochastic robust approximation crosstalk cancellation method based on random perturbation matrix modeling the variations of the transfer function matrix. Under the free-field condition, simulation results demonstrate the effectiveness of the proposed method.

关键词： 3-D audio Crosstalk cancellation Stochastic approximation problem Robustness

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Automatic Tongue Tracking in X-Ray Images

引用

Chinese Journal of Electronics 2015年第4期24卷 767-771页

作者： LUO Changwei LI Rui YU Lingyun YU Jun WANG Zengfu Department of Automation University of Science and Technology of China Institute of Intelligent Machines Chinese Academy of Sciences national Engineering laboratory of speech and language information processing University of Science and Technology of China

X-ray imaging is an effective technique to obtain the continuous motions of the vocal tract during speech, and Active appearance model(AAM) is a useful tool to analyze the X-ray images. However, for the task of tongue tracking in X-ray images, the accuracy of AAM fitting is insufficient. AAM aims to minimize the residual error between the model appearance and the input image. It often fails to accurately converge to the true landmarks. To improve the tracking accuracy, we propose a fitting method by combining Constrained local model(CLM) into *** our method, we first combine the objective functions of AAM and CLM into a single objective function. Then,we pro ject out the texture variation and derive a gradient based method to optimize the objective function. Our method effectively incorporates not only the shape prior and global texture, but also local texture around each landmark. Experiments demonstrate that the proposed method significantly reduces the fitting error. We also show that realistic 3D tongue animation can be created by using tongue tracking results of the X-ray images.

关键词： Active appearance model Constrained local model X-ray images Tongue tracking Tongue modeling

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：