检索结果-内蒙古大学图书馆

18th International Forum of Digital Multimedia Communication, IFTC 2021

作者： Zhou, Zhequan Lyu, Shujing Lu, Yue School of Computer Science and Technology East China Normal University Shanghai China School of Communication and Electronic Engineering East China Normal University Shanghai China Shanghai Key Laboratory of Multidimensional Information Processing Shanghai China

ISBN: (纸本)9789811922657

In this paper, we propose a new instance segmentation framework based on a multi-level prediction mechanism, aiming at segmenting overlapped parcels. In this framework, one location on FPN’s feature maps can predict a set of overlapped instances through a multi-level head architecture according to their overlapping order. Besides, to avoid the inherent limit of bounding boxes in object overlapping scenes, our approach is bounding-box free. And we also provide a dataset for overlapped parcel instance segmentation named OLParcel. On a Mask RCNN baseline, our network can improve 4.25% AP, 4.26% recall, and 7.11% MR- 2 on our dataset. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Short-Term Smart Grid Load Forecasting Based on CNN-BiLSTM with Attention Mechanism

Short-Term Smart Grid Load Forecasting Based on CNN-BiLSTM w...

引用

Energy, Power and Grid (ICEPG), International Conference on

作者： Feihu Sun Yanru Zhong School of Computer Science and Information Security Guilin University of Electronic Technology Guilin Guangxi China Guangxi Key Laboratory of Image and Graphic Intelligent Processing Guilin University of Electronic Technology Guilin Guangxi China

ISBN: (数字)9798350377798

ISBN: (纸本)9798350377804

The existing power load data exhibits complex multidimensionality, while existing techniques often focus on learning and forecasting from single load data. To fully exploit the complex multidimensionality of the existing power load data and enhance the accuracy of short-term load forecasting, a smart grid short-term load forecasting model based on attention mechanism and CNN-BiLSTM is proposed. Firstly, the one-dimensional convolutional network (1D-CNN) slides along the temporal dimension to extract multidimensional features such as temperature and humidity from the raw power load data. Secondly, the Bidirectional Long Short-Term Memory network (BiLSTM) captures both historical and future information of the data through gate mechanisms and the bidirectional network structure. Lastly, the attention mechanism allocates weights to focus on important parts of the load sequence data, enhancing the model’s ability to identify key information and further improving the accuracy of short-term power load forecasting. The method is experimented on a real-world multivariate load dataset and compared with existing learning methods. The proposed model achieved lower RMSE in 24-hour load forecasting compared to CNN-LSTM, CNN-BiLSTM, LSTM-Attention, BiLSTM-Attention and CNN-LSTM-Attention models by 13.2%, 10.6%, 6.7%, 5.7% and 5.9% respectively.

关键词： Attention mechanisms Accuracy Load forecasting Electricity Predictive models Feature extraction Data models Smart grids Convolutional neural networks Load modeling

来源：评论

学校读者我要写书评

暂无评论

Optimizing Model-Driven Federated Learning for Rational and Data-Efficient in Social Mobile Network

引用

IEEE Network 2024年

作者： Lu, Jianfeng Zhang, Ying Cao, Shuqin Wang, Wei Tang, Changbing Wuhan University of Science and Technology School of Computer Science and Technology Wuhan430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology China Key Laboratory of Social Computing and Cognitive Intelligence Dalian University of Technology Ministry of Education China Zhejiang Normal University College of Physics and Electronic Information Engineering Jinhua321004 China

With the rise of data-intensive services in mobile networks, the demand for intelligent and efficient learning frameworks has grown exponentially. Federated Learning (FL), as an emerging paradigm, enables decentralized edge devices to collaboratively train large models without sharing raw data, ensuring privacy preservation. However, deploying FL in resource-constrained mobile environments faces significant challenges, such as free-riding and malicious behavior. These challenges hinder model performance and raise concerns about data integrity and inter-user cooperation. To address these issues, we propose a layered dynamic framework for social mobile networks, called FedSMN, which aims to enhance the generalization ability of collaboratively trained large models in FL environments. FedSMN leverages Bayesian inference to evaluate user behavior and employs evolutionary game theory to simulate dynamic interactions, thereby optimizing cooperation strategies and reducing communication overhead. Unlike traditional approaches that assume fully rational actors, FedSMN accounts for irrational behavior and fluctuating resources, designing an efficient incentive mechanism to improve data utilization. Extensive experiments on large-scale datasets demonstrate that FedSMN increases model accuracy by 21% and reduces communication overhead by 49%, paving the way for scalable, intelligent mobile networks powered by large models. © 1986-2012 IEEE.

关键词： Collaborative learning

来源：评论

学校读者我要写书评

暂无评论

Image segmentation with spatiotemporal relationship-guided graph convolution network

引用

Computing 2025年第7期107卷 1-26页

作者： You, Hongfeng Yu, Long Lin, Jianchu Chen, Xiaobing Wang, Wenhao Yu, Kun Wang, Jiacheng Huaiyin Institute of Technology Huaian China Key Laboratory of Software Engineering Technology Xinjiang University Urumuqi China Jiangsu Key Lab of Image and Video Understanding for Social Security and Key Lab of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education Nanjing University of Science and Technology Nanjing China Faculty of Computer and Software Engineering Huaiyin Institute of Technology Huaian China

The size of the receptive field limits traditional deep convolutional networks and needs help to fully capture the detailed semantic information of targets in complex remote sensing images. At the same time, the attributes of boundary pixels are ambiguous due to many external factors, such as illumination and imaging technology. To solve these problems, we propose a spatiotemporal relationship-guided graph convolutional network for image segmentation. We aim to explore the global and contextual semantics of targets in images and judge the attribute information of boundary pixels. On the one hand, a particular contextual attention module is embedded in the feature extractor to improve the feature extractor’s global and local detail representation. On the other hand, the graph convolution layer, enhanced by the recursive network, aggregates the high-level and low-level features of nodes to strengthen the interactivity between semantic information at different levels. At the same time, long-term dependencies are established between boundary pixels and target center pixels to constrain boundary pixels and improve their attribute decision-making ability. It is worth noting that we designed a weighted loss function to supervise, adjust, and optimize different modules separately. Finally, experimental results on the indian driving dataset, uestc all-day scenery datasets and deepglobe land cover classification challenge datasets show that the proposed framework is robust and has good segmentation performance, with F1 of 89.03%, 98.38% and 85.46%, respectively.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CoUDA: Coherence Evaluation via Unified Data Augmentation

arXiv

引用

arXiv 2024年

作者： Zhu, Dawei Wu, Wenhao Song, Yifan Zhu, Fangwei Cao, Ziqiang Li, Sujian School of Computer Science Peking University China National Key Laboratory for Multimedia Information Processing Peking University China Institute of Artificial Intelligence Soochow University China Jiangsu Collaborative Innovation Center for Language Ability Jiangsu Normal University China

Coherence evaluation aims to assess the organization and structure of a discourse, which remains challenging even in the era of large language models. Due to the scarcity of annotated data, data augmentation is commonly used for training coherence evaluation models. However, previous augmentations for this task primarily rely on heuristic rules, lacking designing criteria as guidance. In this paper, we take inspiration from linguistic theory of discourse structure, and propose a data augmentation framework named COUDA. COUDA breaks down discourse coherence into global and local aspects, and designs augmentation strategies for both aspects, respectively. Especially for local coherence, we propose a novel generative strategy for constructing augmentation samples, which involves post-pretraining a generative model and applying two controlling mechanisms to control the difficulty of generated samples. During inference, COUDA also jointly evaluates both global and local aspects to comprehensively assess the overall coherence of a discourse. Extensive experiments in coherence evaluation show that, with only 233M parameters, COUDA achieves state-ofthe-art performance in both pointwise scoring and pairwise ranking tasks, even surpassing recent GPT-3.5 and GPT-4 based metrics. © 2024, CC BY.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Escalator passenger posture anomaly detection based on improved SSD model 12

Escalator passenger posture anomaly detection based on impro...

引用

12th International Conference on information technology in Medicine and Education, ITME 2022

作者： Xu, Teng Chen, Jian Li, Zuoyong Cai, Yuanzheng Fujian University of Technology School of Electronic Electrical Engineering and Physics Fuzhou350118 China Minjiang University Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Control Engineering Fuzhou350121 China

ISBN: (纸本)9798350310153

With the widespread use of escalators in daily life, escalator-related injury factors are threatening public health. To secure escalator system, most of existing studies focus on post hoc statistics and trend to fix these deficiencies to reduce escalator-related potential risks. However, they are not designed for early-stage detection of impending escalator unsafe factors. To solve this problem, this paper proposes an improved Single Shot MultiBox Detector (SSD) algorithm for anomaly detection of passenger posture on escalators. Firstly, we replace the widely used VGG16 backbone of SSD network by simpler MobileNetV2 to accelerate model inference. Secondly, the distance intersection over Union (DIoU) loss function is introduced to further improve performance. Based on the characteristics of IoU and considering the shortcomings of GIoU, DIoU directly regresses the Euclidean distance between the center points of the two boxes to accelerate the convergence. The experimental results show that the improved SSD escalator passenger posture anomaly detection algorithm can efficiently identify the dangerous posture of escalator passengers and start relevant safety protection measures. According to inference of model, the escalator system can be controlled in practice to formulate the corresponding safety protection measures, and improve the safety performance of the escalator operation. © 2022 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Bidirectional Logits Tree: Pursuing Granularity Reconcilement in Fine-Grained Classification

arXiv

引用

arXiv 2024年

作者： Lu, Zhiguang Xu, Qianqian Bao, Shilong Yang, Zhiyong Huang, Qingming Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China School of Computer Science and Technology University of Chinese Academy of Sciences China Key Laboratory of Big Data Mining and Knowledge Management University of Chinese Academy of Sciences China

This paper addresses the challenge of Granularity Competition in fine-grained classification tasks, which arises due to the semantic gap between multi-granularity labels. Existing approaches typically develop independent hierarchy-aware models based on shared features extracted from a common base encoder. However, because coarse-grained levels are inherently easier to learn than finer ones, the base encoder tends to prioritize coarse feature abstractions, which impedes the learning of fine-grained features. To overcome this challenge, we propose a novel framework called the Bidirectional Logits Tree (BiLT) for Granularity Reconcilement. The key idea is to develop classifiers sequentially from the finest to the coarsest granularities, rather than parallelly constructing a set of classifiers based on the same input features. In this setup, the outputs of finer-grained classifiers serve as inputs for coarser-grained ones, facilitating the flow of hierarchical semantic information across different granularities. On top of this, we further introduce an Adaptive Intra-Granularity Difference Learning (AIGDL) approach to uncover subtle semantic differences between classes within the same granularity. Extensive experiments demonstrate the effectiveness of our proposed method. Copyright © 2024, The Authors. All rights reserved.

关键词： Coarse-grained modeling

来源：评论

学校读者我要写书评

暂无评论

ADS Modeling and Safety Verification in Extreme Driving Scenarios

ADS Modeling and Safety Verification in Extreme Driving Scen...

引用

IEEE International Conference on Software Quality, Reliability and Security Companion (QRS-C)

作者： Ying Zhao Yi Zhu Li Zhao Zexin Li Chan Yin School of Computer Science and Technology Jiangsu Normal University Xuzhou China Key Laboratory of Safety-Critical Software Ministry of Industry and Information Technology Nanjing University of Aeronautics and Astronautics Nanjing China

ISBN: (数字)9798350365658

ISBN: (纸本)9798350365665

Autonomous vehicles play a vital role in improving traffic efficiency and reducing traffic accidents. The spatio-temporal data in driving scenarios is complex and unpredictable, with uncertain interaction behavior. Despite efforts by developers to model road dynamics for evaluating Autonomous Driving Systems (ADS), safety concerns under accidental and catastrophic road conditions have been overlooked. Firstly, to tackle this, this paper proposes extending feature variables for catastrophic scenarios using Domain-Specific Modeling Language (DSML). Introducing the Accidental and Catastrophic Automatic Driving Scenario Modeling Language (AC-ADSML) aims to accurately model ADS behavior in accidental and catastrophic road conditions. The AC-ADSML meta-model categorizes scenario variables and specifies abstract and concrete syntax along with operational semantics. Secondly, it is then transformed into Networks of Stochastic Hybrid Automata (NSHA) following ATL rules. The CARLA simulator is employed to replicate urban driving scenarios during natural disasters with PCTL scenario specifications. Model security can be verified using UPP AAL-SMC. Finally, an autonomous vehicle with adaptive cruise control is used as an example to verify the safety of the proposed model under different road condition parameters.

关键词： Analytical models Adaptation models Computational modeling Roads Disasters Semantics Syntactics Security Autonomous vehicles Meteorology

来源：评论

学校读者我要写书评

暂无评论

Dynamic Uncertainty Estimation for Offline Reinforcement Learning 39

Dynamic Uncertainty Estimation for Offline Reinforcement Lea...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Wang, Jiesheng Li, Lin Wei, Wei Zhang, Yujia Yang, Xin Key Laboratory of Computational Intelligence and Chinese Information Processing of Ministry of Education School of Computer and Information Technology Shanxi University Shanxi Taiyuan China School of Computing and Artificial Intelligence Southwestern University of Finance and Economics Sichuan Chengdu China

ISBN: (纸本)157735897X

Offline reinforcement learning confronts the distributional shift challenge, a consequence of learning policy from static datasets. Current methods primarily handle this issue by aligning the learned policy with the behavior policy or conservatively estimating Q-values for out-of-distribution (OOD) actions. However, these approaches can lead to overly pessimistic estimation of Q-values of the OOD actions in unfamiliar situations, resulting in a suboptimal policy. To address this, we propose a new method, Dynamic Uncertainty estimation for Offline Reinforcement Learning. This method introduces a base density-truncated OOD data sampling approach to reduce the impact of extrapolation errors on uncertainty estimation. It enables conservative estimation of Q-values for OOD actions while avoiding negative impacts on in-distribution data. We also develop a dynamic uncertainty estimation mechanism to prevent excessive pessimism and enhance the generalization of the Q-function. This mechanism dynamically adjusts the degree of pessimism in the Q-function by minimizing the error between target and estimated values. Our method outperforms existing algorithms, as demonstrated by experimental results based on the D4RL benchmark, and proves its superiority in addressing the distributional shift challenge. © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Uncertainty analysis

来源：评论

学校读者我要写书评

暂无评论

Uncertainty-Aware Dynamic Learning for Cross-Domain Few-Shot Scene Classification from Remote Sensing Imagery

Uncertainty-Aware Dynamic Learning for Cross-Domain Few-Shot...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Can Li He Chen Yin Zhuang Shanghang Zhang Beijing Key Laboratory of Embedded Real-Time Information Processing Technology Beijing Institute of Technology Beijing China School of Computer Science Peking University Beijing China

Cross-domain few-shot scene classification (CDFSSC) is devoted to transferring knowledge from the source domain to the target domain and facilitating few-shot classification for the target domain. However, due to the domain shifts between source and target domains, high uncertainty would be generated in the knowledge transfer process, leading to unreliable cross-domain learning, which degenerates classification performance on the target domain severely. Thus, in this paper, aiming to reduce the interference of high uncertainty and improve the reliability of cross-domain knowledge transfer, a novel uncertainty-aware dynamic learning (UDL) framework is proposed for CDFSSC from remote sensing imagery. First, a mean-teacher architecture combining pseudo-labeling and consistency regularization is utilized to achieve cross-domain learning. Second, a UDL strategy is proposed to divide data into positive and negative samples based on a well-designed uncertainty-aware dynamic threshold, conducting positive and negative learning respectively, to advance a more reliable knowledge transfer. Third, to further improve cross-domain capability, a self-entropy loss is designed to reduce the epistemic uncertainty of the model. Extensive experiment results indicate the superiority of our proposed methods.

关键词：

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：