检索结果-内蒙古大学图书馆

2nd International Conference on Digital Society and Intelligent Systems, DSInS 2022

作者： Li, Yiming Chen, Jianpeng Ren, Yazhou Pu, Xiaorong School of Computer Science and Engineering Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China China

ISBN: (纸本)9781510663244

During the COVID-19 coronavirus epidemic, people usually wear masks to prevent the spread of the virus, which has become a major obstacle when we use face-based computer vision techniques such as face recognition and face detection. So masked face inpainting technique is desired. Actually, the distribution of face features is strongly correlated with each other, but existing inpainting methods typically ignore the relationship between face feature distributions. To address this issue, in this paper, we first show that the face image inpainting task can be seen as a distribution alignment between face features in damaged and valid regions, and style transfer is a distribution alignment process. Based on this theory, we propose a novel face inpainting model considering the probability distribution between face features, namely Face Style Self-Transfer Network (FaST-Net). Through the proposed style self-transfer mechanism, FaST-Net can align the style distribution of features in the inpainting region with the style distribution of features in the valid region of a face. Ablation studies have validated the effectiveness of FaST-Net, and experimental results on two popular human face datasets (CelebA and VGGFace) exhibit its superior performance compared with existing state-of-the-art methods. © 2023 SPIE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

LIDAR:learning from imperfect demonstrations with advantage rectification

引用

Frontiers of computer science 2022年第1期16卷 57-66页

作者： Xiaoqin ZHANG Huimin MA Xiong LUO Jian YUAN Department of EE Tsinghua UniversityBeijing100084China School of Computer and Communication Engineering University of Science and Technology BeijingBeijing100083China

In actor-critic reinforcement learning(RL)algorithms,function estimation errors are known to cause ineffective random exploration at the beginning of training,and lead to overestimated value estimates and suboptimal *** this paper,we address the problem by executing advantage rectification with imperfect demonstrations,thus reducing the function estimation *** with expert demonstrations has been widely adopted to accelerate the learning process of deep reinforcement learning when simulations are expensive to ***,existing methods,such as behavior cloning,often assume the demonstrations contain other information or labels with regard to performances,such as optimal assumption,which is usually incorrect and useless in the real *** this paper,we explicitly handle imperfect demonstrations within the actor-critic RL frameworks,and propose a new method called learning from imperfect demonstrations with advantage rectification(LIDAR).LIDAR utilizes a rectified loss function to merely learn from selective demonstrations,which is derived from a minimal assumption that the demonstrating policies have better performances than our current *** learns from contradictions caused by estimation errors,and in turn reduces estimation *** apply LIDAR to three popular actor-critic algorithms,DDPG,TD3 and SAC,and experiments show that our method can observably reduce the function estimation errors,effectively leverage demonstrations far from the optimal,and outperform state-of-the-art baselines consistently in all the scenarios.

关键词： learning from demonstrations actor-critic reinforcement learning advantage rectification

来源：评论

学校读者我要写书评

暂无评论

Taking LM as the Brain: A Novel Approach Integrating Language Model and Generative Agent for Intelligent Decision Systems

引用

International Journal of Network Security 2024年第3期26卷 386-393页

作者： Yang, Jingkang Cai, Xiaodong Liu, Yining Chen, Mingyao Chang, Chin-Chen School of Information and Communication Guilin University of Electronic Technology Guilin541004 China School of Computer Science and Information Security Guilin University of Electronic Technology Guilin541004 China Research and Development Department Guilin Topintelligent Communication Technology Company Limited Guilin541004 China Department of Information Engineering and Computer Science Feng Chia University Taichung 407102 Taiwan

The Large Language Model (LLM) has demonstrated significant capabilities in intelligent robotics and Autonomous Driving(AD). Compared to traditional end-to-end models, decision reasoning in the form of language exhibits enhanced generalization and interpretability. To harness the inferential decision-making capabilities of large models more effectively, we propose a generative intelligent agent capable of translating various sensor signals into language descriptions, thereby utilizing Language Models (LMs) for decision-making. Furthermore, to address the practical deployment challenges of large models, we fine-tuned a GPT-2 model on an intelligent traffic corpus. The fine-tuned model performs competitively with LLMs in the domain of traffic decision-making. This approach can potentially deploy LMs directly on computationally constrained in-vehicle platforms to reduce communication latency. We evaluated the proposed method on a simulated dataset, and the results indicate that the pure language decision model surpasses the vector-based and fusion models. Moreover, our method exhibits significantly higher efficiency than LLMs. © (2024), (International Journal of Network Security). All rights reserved.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

MCFTNet: Multimodal Cross-Layer Fusion Transformer Network for Hyperspectral and LiDAR Data Classification

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 12803-12818页

作者： Huang, Wei Wu, Tianren Zhang, Xueyu Li, Liangliang Lv, Ming Jia, Zhenhong Zhao, Xiaobin Ma, Hongbing Vivone, Gemine Xinjiang University School of Computer Science and Technology Urumqi830017 China Xinjiang University Key Laboratory of Signal Detection and Processing Urumqi830017 China Guangxi University School of Computer and Electronic Information Nanning530004 China Guangxi University Guangxi Key Laboratory of Multimedia Communications and Network Technology Nanning530004 China Beijing Institute of Technology School of Information and Electronics Beijing100081 China University of Science and Technology Beijing School of Computer and Communication Engineering Beijing100083 China Tsinghua University Department of Electronic Engineering Beijing100084 China National Research Council Tito85050 Italy National Biodiversity Future Center NBFC Palermo90133 Italy

Remote sensing image classification is a popular yet challenging field. Many researchers have combined convolutional neural networks (CNNs) and Transformers for hyperspectral imaging (HSI) classification tasks. However, in traditional Transformers, shallow-level information does not propagate well to deeper layers, which can lead to spatial variations and overfitting. Moreover, traditional Transformer models use an external classification token (CLS token) that is randomly initialized and often struggles to generalize effectively. In this article, we combine the strengths of HSI and light detection and ranging (LiDAR) data, using LiDAR as an external CLS token, which significantly enhances classification accuracy and reliability. We propose a new multimodal cross-layer fusion transformer network (MCFTNet), integrating CNNs with the latest Transformer networks. It includes a CNN for extracting spatial features and a hybrid cross-patch attention mechanism for land cover classification, leveraging LiDAR data to generate CLS and HSI patch tokens. More importantly, to reduce the loss of valuable information during the layer-by-layer propagation, we designed cross-layer skip connections. Through adaptive learning, these cross-layer fusions help address the gradient vanishing problem in deep networks while preserving early-layer features. This enables the model to better integrate information from different layers, enhancing both its stability and performance. We carried out in-depth experiments on commonly used benchmark datasets, specifically, the university of Houston dataset, the Trento dataset, and the university of Southern Mississippi Gulf Park dataset. We compared the results of the proposed MCFTNet model with those obtained from state-of-the-art Transformer models, classical CNNs, and traditional classifiers. As a result, the MCFTNet model outshones them all in terms of performance. © 2008-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

LOPO: An Out-of-order Layer Pulling Orchestration Strategy for Fast Microservice Startup 42

LOPO: An Out-of-order Layer Pulling Orchestration Strategy f...

引用

42nd IEEE International Conference on computer Communications, INFOCOM 2023

作者： Gu, Lin Huang, Junhao Huang, Shaoxing Zeng, Deze Li, Bo Jin, Hai Huazhong University of Science and Technology National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Lab Cluster and Grid Computing Lab School of Computer Science and Technology Wuhan China China University of Geosciences School of Computer Science Wuhan China Hong Kong University of Science and Technology Department of Computer Science and Engineering Hong Kong

ISBN: (纸本)9798350334142

Container based microservices have been widely applied to promote the cloud elasticity. The mainstream Docker containers are structured in layers, which are organized in stack with bottom-up dependency. To start a microservice, the required layers are pulled from a remote registry and stored on its host server, following the layer dependency order. This incurs long microservice startup time and hinders the performance efficiency. In this paper, we discover that, for the first time, the layer pulling order can be adjusted to accelerate the microservice startup. Specifically, we address the problem on microservice layer pulling orchestration for startup time minimization and prove it as NP-hard. We propose a Longest-chain based Out-of-order layer Pulling Orchestration (LOPO) strategy with low computational complexity and guaranteed approximation ratio. Through extensive real-world trace driven experiments, we verify the efficiency of our LOPO and demonstrate that it reduces the microservice startup time by 22.71% on average in comparison with state-of-the-art solutions. © 2023 IEEE.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

Incipient Fault Detection of CRH Suspension system Based on PRPCA and Wasserstein Distance 42

Incipient Fault Detection of CRH Suspension system Based on ...

引用

42nd Chinese Control Conference, CCC 2023

作者： Fang, Kangyue Wu, Yunkai Zhou, Yang Zhu, Zhiyu Zeng, Qingjun College of Automation Jiangsu University of Science and Technology Zhenjiang212100 China School of Computer Science and Engineering Jiangsu University of Science and Technology Zhenjiang212100 China

ISBN: (纸本)9789887581543

As an important part of CRH(China Railway High-speed) trains, the stability and stationarity of a suspension system is of great significance to the vehicle system. Based on the framework of probability relevant principal component analysis(PRPCA), a novel data-driven based incipient fault detection method is proposed. Firstly, simulation data including fault information is derived from Simpack-Matlab/Simulink co-simulation platform. Secondly, the real-time monitoring of high-speed train suspension system is proposed based on PRPCA theory combined with wasserstein distance. Furthermore, compared with the traditional PCA based fault detection and diagnosis (FDD) methods, the proposed PRPCA-based method has a better performance and is more suitable for actual fault data has nonlinear and non-Gaussian characteristics. Finally, according to the comparison results with other multivariate statistical analysis based methods, the incipient fault detection method proposed in this paper has a higher sensitivity to the incipient spring/damping faults of CRH suspension system. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Using Recurrent Neural Networks to Forecast Climate Change: A Time Series Analysis of Global Temperature Variability 1

Using Recurrent Neural Networks to Forecast Climate Change: ...

引用

1st International Conference on Advances in computer science, Electrical, Electronics, and Communication Technologies, CE2CT 2025

作者： Rathore, Nitasha Anubha Singh, Kuljeet Kaur, Gaganpreet Sharma, Chetna Sinha, Anurag Bharatiya Vidyapeeth College of Engineering Department of Computer Science and Engineering New Delhi India KIET Group of Institutions India School of Sciences Christ University Department of Computer Science Delhi NCR India Chitkara University Institute of Engineering and Technology Chitkara University Punjab India IGNOU Department of Computer science New Delhi India

ISBN: (纸本)9798331518578

Predicting the upcoming weather instances is very crucial. It depends on different climatic parameters like humidity, pressure, temperature, etc. In this paper, the historical data of the weather in the India area is used for future weather instances of the India for farmers' convenience in terms of the agricultural instance which depends on the weather and, functioning according to that which will restore the energy. For weather forecasting we have use the machine learning algorithm and probabilistic predictions of the predictive analytics based on soft computing and. NGBoost algorithm and. Linear models of the machine learning for predictive. Comparative weather incidence spaced on the historical data. We have also used, sliding window algorithm of the statistics for predicting the ideology of the concept of different contrasted windows and year. We've also used utility thirds and machine learning algorithm, classified for predicting the weather based on different features The overall implementation in this paper, shows the accuracy which we have gathered from the data set. An implementation of algorithm which ranges between 80% to 90% and the entire algorithm have been compared based on the feature instances. Work can be concluded on the measurement of the algorithm, which we have got after the implementation of Models. Which rely upon the different data features and thus it can be beneficial for preserving the energy and materials in the India agriculture area and forecasting the weather as per day Agricultural conditions. © 2025 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

MSAF: A Multi-Scale Attention-Fusion Model for Long-Term Cell-Level Network Traffic Prediction 6

MSAF: A Multi-Scale Attention-Fusion Model for Long-Term Cel...

引用

6th International Conference on Electrical, Control and Instrumentation engineering, ICECIE 2024

作者： Ma, Xuesen Li, Jingqi Li, Dacheng Li, Yangyu Qiao, Yan School of Computer Science and Information Engineering Hefei University of Technology Hefei China Hefei University of Technology Intelligent Interconnected Systems Laboratory of Anhui Province Hefei230009 China Hefei Institutes of Physical Science Chinese Academy of Sciences Hefei China Hefei University of Technology Hefei China

ISBN: (纸本)9798350380040

Implementing precise network traffic prediction is essential for optimizing base station strategies, allocating network resources effectively, reducing costs, and ensuring service quality in network management. Currently, most mainstream forecasting methods primarily focus on either the time domain or the frequency domain, attempting to capture and express time series data features from a one-dimensional perspective. This one-dimensional approach can be limiting, as relying solely on a single dimension may lead to incomplete or suboptimal modeling, it fails to fully capture the intricate interactions between time-domain and frequency-domain that define network traffic behaviors. To address this, we propose a novel Multi-Scale Attention-Fusion (MSAF) model. The key module in MSAF, named Attention-Fusion, integrates both time-domain and frequency-domain features of network traffic data. This effectively addresses the limitations of previous attention mechanisms in simultaneously identifying frequency components and temporal patterns, providing enhanced modeling capabilities for complex network traffic data. Additionally, the Attention-Fusion module utilizes a DSConv module combined with the attention mechanism to identify correlations between sequences of different lengths. Evaluation of MSAF on two real-world network traffic datasets demonstrates that MSAF outperforms other mainstream methods in terms of forecasting performance. © 2024 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Synchronous detection of multiple optical characteristics of atmospheric aerosol by coupled photoacoustic cavity

引用

Chinese Physics B 2022年第6期31卷 278-284页

作者：靳华伟胡仁志谢品华罗平 Key Laboratory of Environmental Optics&Technology Anhui Institute of Optics and Fine MechanicsChinese Academy of SciencesHefei 230031China Institute of Environment Friendly Materials and Occupational Health Anhui University of Science and TechnologyWuhu 241000China State Key Laboratory of Coal Resources in Western China Xi’an University of Science and TechnologyXi’an 710054China Mechanical Engineering School Anhui University of Science and TechnologyHuainan 232001China

Owing to the influence of sampling loss,cavity difference and detecting source,the multi-optical parameter measurement of atmospheric aerosol cannot be detected simultaneously in the same reference *** order to solve this problem,a new method of simultaneously detecting the aerosol optical parameters by coupling cavity ring-down spectrometer with photoacoustic spectroscopy is ***,the coupled photoacoustic cavity is formed by the organic fusion of the photoacoustic cavity and the ring-down ***,the integrated design of the coupling spectroscopy system is carried ***,the extinction coefficient and absorption coefficient of aerosol are measured simultaneously by the system,and then the aerosol scattering coefficient and single albedo are calculated *** accuracy of the system is verified by comparing with the data from the environmental quality monitoring station,which provides a new idea for the detection of multi-optical characteristics of atmospheric aerosol.

关键词： spectroscopy synchronous detection coupling spectrum coupled photoacoustic cavity

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Twitter Feeds Using Flask Environment: A Superior Application of Data Analysis

引用

Annals of Data science 2024年第1期11卷 159-180页

作者： Modi, Astha Shah, Khelan Shah, Shrey Patel, Samir Shah, Manan Department of Information and Communication Technology School of Technology Pandit Deendayal Energy University Gujarat Gandhinagar382426 India Department of Computer Science School of Technology Pandit Deendayal Energy University Gujarat Gandhinagar382426 India Department of Chemical Engineering School of Energy Technology Pandit Deendayal Energy University Gujarat Gandhinagar382426 India

In this challenging world, social media plays a vital role as it is at the pinnacle of data sharing. The advancement in technology has made a huge amount of information available for data analysis and it is on the hotlist nowadays. Opinions of the people are expressed and shared across various social media platforms like Twitter, Facebook, and Instagram. Twitter is a prodigious platform containing an ample amount of data and analyzing the data is of topmost priority. One of the most widely utilized approaches for classifying an individual’s emotions displayed in subjective data is sentiment analysis. Sentiment analysis is done using various algorithms of machine learning like Support Vector Machine, Naive Bayes, Long Short-Term Memory, Decision Tree Classifier, and many more, but this paper aims at the generalized way of performing Twitter sentiment analysis using flask environment. Flask environment provides various inbuilt functionalities to analyze the sentiments of text into three different categories: positive, negative, and neutral. Also, it makes API calls to the Twitter Developer account to fetch the Twitter data. After fetching and analyzing the data, the results get displayed on a webpage containing the percentage of positive, negative, and neutral tweets for a phrase in a pie chart. It displays the language analysis for the same phrase. Furthermore, the webpage calls attention to the tweets done on that phrase and reveals the details of the tweets. Considering the major industry runners of three different sectors namely Enterprises, Sports Apparel Industry, and Multimedia Industry, we have analyzed and compared sentiments of two different Multinational companies from each sector. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022.

关键词： Python

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：