检索结果-内蒙古大学图书馆

11th International Conference on Information Technology and Quantitative Management, ITQM 2024

作者： Zhu, Siyu Zhou, Fenfen Li, Kai Wen, Haonan Liu, Minghao Computer School Beijing Information Science and Technology University 102206 No.55 Taihang RoadChangping District Beijing China Beijing Key Laboratory of Information Service Engineering Beijing Union University Beijing100101 China College of Robotics Beijing Union University Beijing100027 China National Engineering Research Center of Visual Technology School of Computer Science Peking University Beijing100871 China Research Center on Fictitious Economy and Data Science Chinese Academy of Sciences No. 80 of Zhongguancun East Road Haidian District Beijing100190 China Yuquan Road Shijingshan District Beijing100049 China

The traditional domain adaptation task is not sufficient in mining inter domain image context. We propose a solution from the perspective of image semantic level retrieval. Image semantic level retrieval is based on the high-level semantics of images, retrieving the similarity or correspondence between images in terms of content and scene between domains. Our work focuses on the meaning, structure, and contextual information expressed by the target image, and retrieves the most relevant image from the source domain to improve the accuracy of domain adaptation. Furthermore, in response to the negative transfer brought about by image retrieval, we will introduce a category level weight adjustment module to further improve the accuracy of semantic level retrieval and weaken the negative impact caused by low precision categories. Finally, we demonstrated the effectiveness of semantic level image retrieval through experiments. © 2024 The Authors.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Better Language Models of Code through Self-Improvement

arXiv

引用

arXiv 2023年

作者： To, Hung Quoc Bui, Nghi D.Q. Guo, Jin Nguyen, Tien N. FPT Software AI Center Department of Computer Science Fulbright University Viet Nam Viet Nam School of Computer Science McGill University Canada Mila - Quebec AI Institute Canada School of Engineering and Computer Science The University of Texas Dallas United States

Pre-trained language models for code (PLMCs) have gained attention in recent research. These models are pre-trained on large-scale datasets using multi-modal objectives. However, fine-tuning them requires extensive supervision and is limited by the size of the dataset provided. We aim to improve this issue by proposing a data augmentation framework using knowledge distillation. Our framework utilizes knowledge gained during the pre-training and fine-tuning stage to generate pseudo data, which is then used as training data for the next step. We incorporate this framework into the state-of-the-art language models, such as CodeT5, CodeBERT, and UnixCoder. The results show that our framework significantly improves PLMCs’ performance in sequence-generation tasks, such as code summarization and code generation in the CodeXGLUE benchmark. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

FFCI: A Camera and IMU Sensors Based Multi-modal Neural Network for Activity Recognition in Smart Factory

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Wang, Yujue Niu, Xin Lv, Xianwei Yu, Chen 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 Wuhan430074 China

Worker activity recognition is an important aspect of the construction of smart factory. The development of deep neural networks and the widespread distribution of sensors in the smart factory have brought opportunities for the recognition of workers' activities. The existing methods based on camera and IMU sensors respectively have problems of visual occlusion and difficulty in recognition of similar activities, which result in the reduction of recognition accuracy. Therefore, we propose a feature-level fusion based on camera and IMU sensor (FFCI) for activity recognition in smart factory. The FFCI employs an optimized multi-modal fusion strategy to integrate the visual information and the action information of workers' activities. In order to verify the effectiveness of FFCI, we define a simple yet fine-grained assembly task and collect six common operations. Furthermore, we add visual occlusion and similar activity data to get closer to the smart factory. Finally, we evaluate the FFCI on the collected dataset and achieve a recognition accuracy of 98.17%, demonstrating its effectiveness in accurately classifying workers' activities. © 1975-2011 IEEE.

关键词： Factory automation

来源：评论

学校读者我要写书评

暂无评论

ASGEA: Exploiting Logic Rules from Align-Subgraphs for Entity Alignment

arXiv

引用

arXiv 2024年

作者： Luo, Yangyifei Chen, Zhuo Guo, Lingbing Li, Qian Zeng, Wenxuan Cai, Zhixin Li, Jianxin School of Computer Science and Engineering Beihang University China College of Computer Science and Technology Zhejiang University China Bejing Advanced Innovation Center for Big Data and Brain Computing Beihang University China

Entity alignment (EA) aims to identify entities across different knowledge graphs that represent the same real-world objects. Recent embedding-based EA methods have achieved state-of-the-art performance in EA yet faced interpretability challenges as they purely rely on the embedding distance and neglect the logic rules behind a pair of aligned entities. In this paper, we propose the Align-Subgraph Entity Alignment (ASGEA) framework to exploit logic rules from Align-Subgraphs. ASGEA uses anchor links as bridges to construct Align-Subgraphs and spreads along the paths across KGs, which distinguishes it from the embedding-based methods. Furthermore, we design an interpretable Path-based Graph Neural Network, ASGNN, to effectively identify and integrate the logic rules across KGs. We also introduce a node-level multi-modal attention mechanism coupled with multi-modal enriched anchors to augment the Align-Subgraph. Our experimental results demonstrate the superior performance of ASGEA over the existing embedding-based methods in both EA and Multi-Modal EA (MMEA) tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： computer circuits

来源：评论

学校读者我要写书评

暂无评论

Reveal training performance mystery between Tensor Flow and PyTorch in the single GPU environment

引用

science China(Information sciences) 2022年第1期65卷 147-163页

作者： Hulin Dai Xuan PENG Xuanhua SHI Ligang HE Qian XIONG Hai JIN National Engineering Research Center for Big Data Technology and System Service Computing Technology and System LabSchool of Computer Science and Technology Huazhong University of Science and Technology Department of Computer Science University of Warwick

Deep learning has gained tremendous success in various fields while training deep neural networks(DNNs) is very compute-intensive, which results in numerous deep learning frameworks that aim to offer better usability and higher performance to deep learning practitioners. Tensor Flow and Py Torch are the two most popular frameworks. Tensor Flow is more promising within the industry context, while Py Torch is more appealing in academia. However, these two frameworks differ much owing to the opposite design philosophy:static vs dynamic computation graph. Tensor Flow is regarded as being more performance-friendly as it has more opportunities to perform optimizations with the full view of the computation graph. However, there are also claims that Py Torch is faster than Tensor Flow sometimes, which confuses the end-users on the choice between them. In this paper, we carry out the analytical and experimental analysis to unravel the mystery of comparison in training speed on single-GPU between Tensor Flow and Py Torch. To ensure that our investigation is as comprehensive as possible, we carefully select seven popular neural networks, which cover computer vision, speech recognition, and natural language processing(NLP). The contributions of this work are two-fold. First, we conduct the detailed benchmarking experiments on Tensor Flow and Py Torch and analyze the reasons for their performance difference. This work provides the guidance for the end-users to choose between these two frameworks. Second, we identify some key factors that affect the performance,which can direct the end-users to write their models more efficiently.

关键词： deep learning performance comparison TensorFlow PyTorch

来源：评论

学校读者我要写书评

暂无评论

SPViT: Accelerate Vision Transformer Inference on Mobile Devices via Adaptive Splitting and Offloading

引用

IEEE Transactions on Mobile Computing 2025年

作者： Zhao, Sifan Liu, Tongtong Jin, Hai Yao, Dezhong 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 Wuhan430074 China

The Vision Transformer (ViT), which benefits from utilizing self-attention mechanisms, has demonstrated superior accuracy compared to CNNs. However, due to the expensive computational costs, deploying and inferring ViTs on resource-constrained mobile devices has become a challenge. To resolve this challenge, we conducted an empirical analysis to identify performance bottlenecks in deploying ViTs on mobile devices and explored viable solutions. In this paper, we propose SPViT, an adaptive split and offloading method that accelerates ViT inference on mobile devices. SPViT executes collaborative inference of ViT across available edge devices. We introduce a fine-grained splitting technique for the vision transformer structure. Furthermore, we propose an algorithm based on the Auto Regression model to predict partition latency and adaptive offload partitions. Finally, we design offline and online optimization methods to minimize the computational and communication overhead on each device. Based on real-world prototype experiments, SPViT effectively reduces inference latency by 2.2x to 3.3x across four state-of-the-art models. © 2002-2012 IEEE.

关键词： Regression analysis

来源：评论

学校读者我要写书评

暂无评论

data Poisoning Attack to X-armed Bandits 21

Data Poisoning Attack to X-armed Bandits

引用

21st IEEE International Conference on Trust, Security and Privacy in Computing and Communications, TrustCom 2022

作者： Luo, Zhi Li, Youqi Chen, Lixing Xu, Zichuan Zhou, Pan Huazhong University of Science and Technology School of Cyber Science and Engineering China Beijing Institute of Technology School of Cyberspace Science and Technology School of Computer Science China Shanghai Jiao Tong University Institute of Cyber Science and Technology China Dalian University of Technology School of Software China Huazhong University of Science and Technology Hubei Engineering Research Center on Big Data Security School of Cyber Science and Engineering China

ISBN: (纸本)9781665494250

X-armed bandits have achieved the state-of-the-art performance in optimizing unknown stochastic continuous functions, which can model many machine learning tasks, specially in big data-driven personalized recommendation. However, bandit algorithms are vulnerable to adversarial attacks. Existing works mainly focus on attacking multi-armed bandits in discrete setting;nevertheless, the attacks against X-armed bandits in continuous setting have not been well explored. In this paper, we aim to bridge this gap and investigate the robustness problem for the X-armed bandits. Specifically, we consider data poisoning attack and propose an attack algorithm named Confidence Poisoning Attack algorithm, which could hijack the clean tree-based X-armed bandits algorithm, i.e., high confidence tree (HCT) and make it choose the nodes including the arm targeted by the attacker very frequently with a sub-linear attack cost, i.e., O(Tα)(0 © 2022 IEEE.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Semantic Reasoning Technology on Temporal Knowledge Graph 19th

Semantic Reasoning Technology on Temporal Knowledge Graph

引用

19th International Conference on Web Information Systems and Applications, WISA 2022

作者： Li, Jianuo Zhao, Feng Jin, Hai 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 Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9783031203084

Semantic reasoning techniques based on knowledge graphs have been widely studied since they were proposed. Previous studies are mostly based on closed-world assumptions, which cannot reason about unknown facts. To this end, we propose the Two-Stage Temporal Reasoning Model (TSTR) for reasoning about future facts. In the first stage, probability of future facts occurring is reasoned using repeated information in history. In the second stage, the semantics of the neighborhood nodes are aggregated using the structural encoder and the temporal information is captured using the temporal encoder. The predicted probabilities are obtained by the decoder. Finally, the candidate entity probabilities of the two-stage reasoning are weighted to achieve the prediction of the two-stage fusion. We tested the performance of the TSTR on public datasets and the results demonstrated the effectiveness of the TSTR. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Commonsense Knowledge Construction with Concept and Pretrained Model 19th

Commonsense Knowledge Construction with Concept and Pretrain...

引用

19th International Conference on Web Information Systems and Applications, WISA 2022

作者： Cai, Hanjun Zhao, Feng Jin, Hai 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 Huazhong University of Science and Technology Wuhan China

ISBN: (纸本)9783031203084

Commonsense knowledge (CSK) is the information that people use in daily life but do not often mention. It summarizes the practical knowledge about how the world works. Existing machines have knowledge but lack commonsense because they do not understand and master commonsense knowledge in the same way that humans do. In the latest works, crowdsourcing-based method is costly and has low coverage, knowledge base completion method can highly fit samples, and methods extracted from unstructured data have the defects of low quality. CG &BF is commonsense knowledge construction with a concept-based generator and a BERT-based filter. We utilize semantic search for node matching and entropy encoder for filtering triples with high abstraction. Two algorithms based on concept aggregation and path credibility are proposed to obtain high-quality CSK triples. We subsequently finetuning a BERT to filter incorrect triples. We obtain 500,000 CSK triples based on ConceptNet, which is superior to other construction methods in novelty and quality. In the reading comprehension task, the three-way attention network is selected as the basic model and the knowledge we generate enables the base model to perform better, which proves that the output of CG &BF has higher quality and ease of use. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Complementing Event Streams and RGB Frames for Hand Mesh Reconstruction

arXiv

引用

arXiv 2024年

作者： Jiang, Jianping Zhou, Xinyu Wang, Bingxuan Deng, Xiaoming Xu, Chao Shi, Boxin National Key Laboratory for Multimedia Information Processing School of Computer Science Peking University China National Engineering Research Center of Visual Technology School of Computer Science Peking University China AI Innovation Center School of Computer Science Peking University China National Key Lab of General AI School of Intelligence Science and Technology Peking University China Institute of Software Chinese Academy of Sciences China University of Chinese Academy of Sciences China

Reliable hand mesh reconstruction (HMR) from commonly-used color and depth sensors is challenging especially under scenarios with varied illuminations and fast motions. Event camera is a highly promising alternative for its high dynamic range and dense temporal resolution properties, but it lacks salient texture appearance for hand mesh reconstruction. In this paper, we propose EvRGBHand – the first approach for 3D hand mesh reconstruction with an event camera and an RGB camera compensating for each other. By fusing two modalities of data across time, space, and information dimensions, EvRGBHand can tackle overexposure and motion blur issues in RGB-based HMR and foreground scarcity as well as background overflow issues in event-based HMR. We further propose EvRGBDegrader, which allows our model to generalize effectively in challenging scenes, even when trained solely on standard scenes, thus reducing data acquisition costs. Experiments on real-world data demonstrate that EvRGBHand can effectively solve the challenging issues when using either type of camera alone via retaining the merits of both, and shows the potential of generalization to outdoor scenes and another type of event camera. For code, models, and dataset, please refer to https://***/***/. Copyright © 2024, The Authors. All rights reserved.

关键词： Cameras

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：