检索结果-内蒙古大学图书馆

Multi-Passage Machine Reading Comprehension Through Multi-Task Learning and Dual Verification

IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING 2024年第10期36卷 5280-5293页

作者： Li, Xingyi Cheng, Xiang Xia, Min Ren, Qiyu He, Zhaofeng Su, Sen Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing 100876 Peoples R China Kuaishou Technol Beijing 100085 Peoples R China

Multi-passage machine reading comprehension (MRC) aims to answer a question by multiple passages. Existing multi-passage MRC approaches have shown that employing passages with and without golden answers (i.e., labeled and unlabeled passages) for model training can improve prediction accuracy. However, when using the unlabeled passages, they either incur the wrong labeling problem or treat the labeled and unlabeled passages equally. In addition, they ignore the original passage information to verify the correctness of the answer. In this paper, we present MLDV-MRC, a novel approach for multi-passage MRC via Multi-task Learning and Dual Verification. MLDV-MRC adopts the extract-then-select framework, where an extractor is first used to predict answer candidates, then a selector is used to choose the final answer. For the extractor, we adopt multi-task learning with generative adversarial training to train it by using both labeled and unlabeled passages. To train the extractor by backpropagation, we propose a hybrid method which combines boundary-based and content-based extracting methods to produce the answer candidate set and its representation. For the selector, we propose to leverage both the information from answer candidates and original passages to verify the final answer. In particular, we propose a global-local memory-augmented neural network to build the representations of original passages, which fuses the passage-level information and word-level information. The experimental results on three open-domain QA datasets confirm the effectiveness of our approach.

关键词： Training Multitasking Task analysis labeling Backpropagation Neural networks Data mining Multi-passage machine reading comprehension question answering multi-task learning generative adversarial training dual verification memory-augmented neural network

来源：评论

学校读者我要写书评

暂无评论

An Energy-efficient ISAC Beam Management Scheme in UAV Communications 25

An Energy-efficient ISAC Beam Management Scheme in UAV Commu...

引用

IEEE Wireless Communications and networking Conference (IEEE WCNC)

作者： Bai, Zhishan Zhang, Jiupeng Ouyang, Qiu Yan, Shi Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350303582;9798350303599

With the advantages of high flexibility and rapid deployment, unmanned aerial vehicles (UAVs) are considered as promising wireless access points for efficient emergency communication. However, UAV base stations (UAV-BSs) are characterized by their susceptibility to environmental interference, leading to jitter and prone to beam misalignment. Additionally, the operational time of a UAV-BS is constrained by the power source. To overcome these obstacles, this paper proposes an energy-efficient beam alignment and tracking scheme based on integrated sensing and communication (ISAC) technology. Specifically, beam alignment is performed based on the predicted angles obtained from the extended Kalman filtering (EKF) algorithm. Then, the beamwidth control is realized by changing the number of activated antennas based on the prediction error derived from EKF. To determine the optimal number of activated antennas for maximizing energy efficiency, a block coordinate descent based fractional programming algorithm is designed. Simulation results verify the superiority of the proposed scheme in terms of communication energy efficiency.

关键词： unmanned aerial vehicle integrated sensing and communication beamwidth control energy efficiency

来源：评论

学校读者我要写书评

暂无评论

A Fuzzy Error Based Fine-Tune Method for Spatio-Temporal Recognition Model 6th

A Fuzzy Error Based Fine-Tune Method for Spatio-Temporal Rec...

引用

6th Chinese Conference on Pattern Recognition and Computer Vision (PRCV)

作者： Li, Jiulin Yang, Mengyu Liu, Yang Xi, Gongli Zhang, Lanshan Tian, Ye Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9789819984282;9789819984299

The spatio-temporal convolution model is widely recognized for its effectiveness in predicting action in various fields. This model typically uses video clips as input and employs multiple clips for inference, ultimately deriving a video-level prediction through an aggregation function. However, the model will give a high confidence prediction result, regardless of whether the input clips have sufficient spatiotemporal information to indicate its class or not. The inaccurate high confidence prediction errors can subsequently affect the accuracy of the video-level results. Although the current approach to mitigating this problem involves increasing the number of clips used, it fails to address this problem from its root causes. To solve this issue, we propose a fine-tuning framework based on Fuzzy error loss, aimed at further refining the well-trained spatio-temporal convolution model that relies on dense sampling. By giving a low confidence prediction output for clips with insufficient spatio-temporal information, our framework strives to enhance the accuracy of video-level motion recognition. We conducted extensive experiments on two motion recognition datasets, namely UCF101 and Kinetics-Sounds, to evaluate the effectiveness of our proposed framework. The results indicate a significant improvement in motion recognition accuracy at the video level on both data sets.

关键词： Action recognition Fine-tuning method Spatio-temporal convolution

来源：评论

学校读者我要写书评

暂无评论

Dynamic Beam Allocation Based on Swap Matching Algorithm Between NGSO Constellations 25

Dynamic Beam Allocation Based on Swap Matching Algorithm Bet...

引用

IEEE Wireless Communications and networking Conference (IEEE WCNC)

作者： Zhang, Yifan Wang, Ying Jia, Huaiqi Zhang, Qiuyang Feng, Linqing Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350303582;9798350303599

The Non-geostationary orbit (NGSO) satellite constellation has gained widespread recognition owing to its exceptional low latency and seamless global coverage. However, with the increasing number of countries launching satellites, there has been a surge in the amount of satellites orbiting in low earth orbit, resulting in a growing strain on both frequency and orbital resources. Therefore, the issue of satellite coexistence has become increasingly critical. Moreover, the uneven distribution of terrestrial users imposes higher demands on beam resource management. This paper proposes a beam allocation strategy based on matching theory, which effectively mitigates harmful interference from the main constellation while optimizing the throughput of the minor constellation. The proposed strategy enables on-demand allocation, which maximizes both service quality and resource utilization, ultimately achieving the objective of constellation coexistence. Simulation results demonstrate that the performance of the proposed strategy is significantly superior compared to other link establishment schemes.

关键词： NGSO constellations coexistence dynamic beam allocation interference avoidance

来源：评论

学校读者我要写书评

暂无评论

Profit-Aware Task Allocation in Satellite Computing

Profit-Aware Task Allocation in Satellite Computing

引用

IEEE International Conference on Web Services (IEEE ICWS)

作者： Huang, Jie Xing, Ruolin Ma, Xiao Zhou, Ao Wang, Shangguang Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350368567;9798350368550

The rapid evolution of satellite networks promises to expand global internet service. However, optimizing task allocation for the efficient and sustainable operation of satellite computing presents complex challenges. Existing approaches usually prioritize energy considerations while neglecting economic aspects, which restricts satellite networks from achieving their full economic potential. In this paper, we address this gap by investigating task allocation in satellite computing. Our approach encourages satellites to consistently provide resources and optimizes battery usage, enabling the completion of more tasks and ultimately maximizing profit. The task allocation approach involves two key components: task pricing and task scheduling. Firstly, we introduce a unique task pricing algorithm that adheres to economic properties, establishing a direct link between satellite utilization and financial income, ensuring economically viable satellite operations. Moreover, we develop two distinct task scheduling algorithms tailored for offline and online scenarios, exploiting dynamic programming and reinforcement learning respectively. Extensive simulations demonstrate that our proposed algorithms effectively enhance task completion rates and optimize total satellite profit.

关键词： Satellite computing task allocation task pricing

来源：评论

学校读者我要写书评

暂无评论

Addressing Catastrophic Forgetting in Federated Learning on Resource-Constrained Devices: A Feature Replay Approach 20th

Addressing Catastrophic Forgetting in Federated Learning on ...

引用

20th International Conference on Intelligent Computing (ICIC)

作者： Gao, Zhipeng Zhang, Junyao Yu, Xinlei Beijing Univ Posts & Telecommun State Key Lab Switching & Networking Technol Beijing Peoples R China

ISBN: (纸本)9789819756742;9789819756759

Federated learning is a new type of privacy-preserving distributed machine learning method. Despite its advantages, Federated learning suffers from catastrophic forgetting due to the temporal variability of data on participating devices. While some methods have been proposed to address this issue in federated learning, they often come at the expense of storage space, which is not a practical solution. This paper presents a federated incremental learning framework designed specifically for resource-constrained devices to address catastrophic forgetting. Leveraging a feature generator, knowledge distillation, and dynamic adaptive weight allocation, the framework addresses catastrophic forgetting and accelerates model convergence. Our framework effectively addresses the issue of catastrophic forgetting, even in the context of resource-constrained devices with limited storage. The results demonstrate significant improvements in our framework compared to existing baselines on the CIFAR-10 and CIFAR-100 datasets.

关键词： Federated learning Continual learning Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Semantic Communication-Enabled Wireless Adaptive Panoramic Video Transmission 25

Semantic Communication-Enabled Wireless Adaptive Panoramic V...

引用

IEEE Wireless Communications and networking Conference (IEEE WCNC)

作者： Gao, Haixiao Sun, Mengying Xu, Xiaodong Han, Shujun Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350303582;9798350303599

In this paper, we propose an adaptive panoramic video semantic transmission (APVST) network built on the deep joint source-channel coding (Deep JSCC) structure for the efficient end-to-end transmission of panoramic videos. The proposed APVST network can adaptively extract semantic features of panoramic frames and achieve semantic feature encoding. To achieve high spectral efficiency and save bandwidth, we propose a transmission rate control mechanism for the APVST via the entropy model and the latitude adaptive model. Besides, we take weighted-to-spherically-uniform peak signal-to-noise ratio (WS-PSNR) and weighted-to-spherically-uniform structural similarity (WS-SSIM) as distortion evaluation metrics, and propose the weight attention module to fuse the weights with the semantic features to achieve better quality of immersive experiences. Finally, we evaluate our proposed scheme on a panoramic video dataset containing 208 panoramic videos. The simulation results show that the APVST can save up to 20% and 50% on channel bandwidth cost compared with other semantic communication-based and traditional video transmission schemes.

关键词： Semantic communication panoramic video latitude adaptive weight attention

来源：评论

学校读者我要写书评

暂无评论

Analysis of Near-Field Effects, Spatial Non-Stationary Characteristics Based on 11-15 GHz Channel Measurement in Indoor Scenario 25

Analysis of Near-Field Effects, Spatial Non-Stationary Chara...

引用

25th IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC)

作者： Miao, Haiyang Tang, Pan Zuo, Weirang Wei, Qi Tian, Lei Zhang, Jianhua Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350393194;9798350393187

In the sixth generation (6G) system, with the further expansion of array element number and frequency bands, wireless communications are expected to operate in the near-field region. The near-field radio communications (NFRC) will become crucial in 6G communication systems. The new mid-band (6-24 GHz) is the 6G potential candidate spectrum. In this paper, we will investigate the channel measurements and characteristics of the emerging NFRC. First, the near-field spherical-wave signal model is derived in detail, and the stationary interval (SI) division method is discussed based on the channel statistical properties. Then, the influence of line-of-sight (LOS) and obstructed-LOS (OLOS) environments on the near-field effects and spatial non-stationary (SnS) characteristics is explored based on the near-field channel measurements at 11-15 GHz band. We hope that this work will give some reference to the NFRC research.

关键词： Near-field new mid-band spherical-wave model spatial non-stationary channel measurements

来源：评论

学校读者我要写书评

暂无评论

Measurement-Based Massive MIMO Channel Characterization in 6 GHz Band for 6G 25

Measurement-Based Massive MIMO Channel Characterization in 6...

引用

IEEE Wireless Communications and networking Conference (IEEE WCNC)

作者： Miao, Haiyang Tang, Pan Zhang, Jianhua Tian, Lei Xu, Huixin Liu, Shuo Gao, Tianyang Li, Yi Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350303582;9798350303599

Massive multiple-input multiple-output (MIMO) is envisioned as a promising technology in 5G-Advanced and sixth-generation (6G) communication. Channel characteristics are important for the development and performance assessment of massive MIMO systems. Besides, the 6 GHz band (5925-7125 MHz) has rapidly become the focus band of industry and academia at present. Therefore, this paper conducts the massive MIMO channel measurements in 6 GHz band in typical urban scenarios, and focuses on the massive MIMO channel characteristics in 6 GHz band. The important channel characteristic parameters such as power delay profile (PDP), delay spread, angular spread, channel capacity and data transmission rate are obtained and investigated, which are statistically different from Third Generation Partnership Project (3GPP) TR 38.901 model parameters. Besides, the impact of bandwidth on the system is important, and the great bandwidth allocation has the ability to provide higher transmission rates for future communication systems. Based on deeply exploration of the channel characteristics in 6 GHz band, this work hopes to provide some reference for the design and optimization of 6G communication systems.

关键词： Massive MIMO 6 GHz band channel measurements channel characterization 6G 3GPP

来源：评论

学校读者我要写书评

暂无评论

keypoint Fusion for RGB-D Based 3D Hand Pose Estimation 38

Keypoint Fusion for RGB-D Based 3D Hand Pose Estimation

引用

38th AAAI Conference on Artificial Intelligence (AAAI) / 36th Conference on Innovative Applications of Artificial Intelligence / 14th Symposium on Educational Advances in Artificial Intelligence

作者： Liu, Xingyu Ren, Pengfei Gao, Yuanyuan Wang, Jingyu Sun, Haifeng Qi, Qi Zhuang, Zirui Liao, Jianxin Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)1577358872

Previous 3D hand pose estimation methods primarily rely on a single modality, either RGB or depth, and the comprehensive utilization of the dual modalities has not been extensively explored. RGB and depth data provide complementary information and thus can be fused to enhance the robustness of 3D hand pose estimation. However, there exist two problems for applying existing fusion methods in 3D hand pose estimation: redundancy of dense feature fusion and ambiguity of visual features. First, pixel-wise feature interactions introduce high computational costs and ineffective calculations of invalid pixels. Second, visual features suffer from ambiguity due to color and texture similarities, as well as depth holes and noise caused by frequent hand movements, which interferes with modeling cross-modal correlations. In this paper, we propose keypoint-Fusion for RGB-D based 3D hand pose estimation, which leverages the unique advantages of dual modalities to mutually eliminate the feature ambiguity, and performs cross-modal feature fusion in a more efficient way. Specifically, we focus cross-modal fusion on sparse yet informative spatial regions (i.e. keypoints). Meanwhile, by explicitly extracting relatively more reliable information as disambiguation evidence, depth modality provides 3D geometric information for RGB feature pixels, and RGB modality complements the precise edge information lost due to the depth noise. keypoint-Fusion achieves state-of-the-art performance on two challenging hand datasets, significantly decreasing the error compared with previous single-modal methods.

关键词： Pixels

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：