检索结果-内蒙古大学图书馆

IEEE Transactions on Intelligent Vehicles 2024年 1-12页

作者： Zhang, Yanan Zhou, Chao Huang, Di State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing China

Existing 3D object detection suffers from expensive annotation costs and poor transferability to unknown data due to the domain gap, Unsupervised Domain Adaptation (UDA) aims to generalize detection models trained in labeled source domains to perform robustly on unexplored target domains, providing a promising solution for cross-domain 3D object detection. Although Self-Training (ST) based cross-domain 3D detection methods with the assistance of pseudo-labeling techniques have achieved remarkable progress, they still face the issue of low-quality pseudo-labels when there are significant domain disparities due to the absence of a process for feature distribution alignment. While Adversarial Learning (AL) based methods can effectively align the feature distributions of the source and target domains, the inability to obtain labels in the target domain forces the adoption of asymmetric optimization losses, resulting in a challenging issue of source domain bias. To overcome these limitations, we propose a novel unsupervised domain adaptation framework for 3D object detection via collaborating ST and AL, dubbed as STAL3D, unleashing the complementary advantages of pseudo labels and feature distribution alignment. Additionally, a Background Suppression Adversarial Learning (BS-AL) module and a Scale Filtering Module (SFM) are designed tailored for 3D cross-domain scenes, effectively alleviating the issues of the large proportion of background interference and source domain size bias. Our STAL3D achieves state-of-the-art performance on multiple cross-domain tasks and even surpasses the Oracle results on Waymo $\rightarrow$ KITTI and Waymo $\rightarrow$ KITTI-rain. IEEE

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

ALSTNet:Autoencoder fused long-and short-term time-series network for the prediction of tunnel structure

引用

Deep Underground Science and Engineering 2025年第1期4卷 72-82页

作者： Bowen Du Haohan Liang Yuhang Wang Junchen Ye Xuyan Tan Weizhong Chen State Key Laboratory of Software Development Environment Beihang UniversityBeijingChina State Key Laboratory of Geomechanics and Geotechnical Engineering Institute of Rock and Soil MechanicsChinese Academy of SciencesWuhanChina

It is crucial to predict future mechanical behaviors for the prevention of structural *** for underground construction,the structural mechanical behaviors are affected by multiple internal and external factors due to the complex *** that the existing models fail to take into account all the factors and accurate prediction of the multiple time series simultaneously is difficult using these models,this study proposed an improved prediction model through the autoencoder fused long-and short-term time-series network driven by the mass number of monitoring ***,the proposed model was formalized on multiple time series of strain monitoring ***,the discussion analysis with a classical baseline and an ablation experiment was conducted to verify the effectiveness of the prediction *** the results indicate,the proposed model shows obvious superiority in predicting the future mechanical behaviors of *** a case study,the presented model was applied to the Nanjing Dinghuaimen tunnel to predict the stain variation on a different time scale in the future.

关键词： autoencoder deep learning structural health monitoring time-series prediction

来源：评论

学校读者我要写书评

暂无评论

swSpAMM:optimizing large-scale sparse approximate matrix multiplication on Sunway Taihulight

引用

Frontiers of Computer Science 2023年第4期17卷 29-41页

作者： Xiaoyan LIU Yi LIU Bohong YIN Hailong YANG Zhongzhi LUAN Depei QIAN State Key Laboratory of Software Development Environment Beijing 100191China School of Computer Science and Engineering Beihang UniversityBeijing 100191China

Although matrix multiplication plays an essential role in a wide range of applications,previous works only focus on optimizing dense or sparse matrix *** Sparse Approximate Matrix Multiply(SpAMM)is an algorithm to accelerate the multiplication of decay matrices,the sparsity of which is between dense and sparse *** addition,large-scale decay matrix multiplication is performed in scientific applications to solve cutting-edge *** optimize large-scale decay matrix multiplication using SpAMM on supercomputers such as Sunway Taihulight,we present swSpAMM,an optimized SpAMM algorithm by adapting the computation characteristics to the architecture features of Sunway ***,we propose both intra-node and inter-node optimizations to accelerate swSpAMM for large-scale *** intra-node optimizations,we explore algorithm parallelization and block-major data layout that are tailored to better utilize the architecture advantage of Sunway *** inter-node optimizations,we propose a matrix organization strategy for better distributing sub-matrices across nodes and a dynamic scheduling strategy for improving load balance across *** compare swSpAMM with the existing GEMM library on a single node as well as large-scale matrix multiplication methods on multiple *** experiment results show that swSpAMM achieves a speedup up to 14.5×and 2.2×when compared to xMath library on a single node and 2D GEMM method on multiple nodes,respectively.

关键词： approximate calculation sunway processor performance optimization

来源：评论

学校读者我要写书评

暂无评论

VGGAN: Visual Grounding GAN Using Panoptic Transformers 8

VGGAN: Visual Grounding GAN Using Panoptic Transformers

引用

8th International Conference on Image, Vision and Computing, ICIVC 2023

作者： Quan, Fengnan Lang, Bo Beihang University State Key Laboratory of Software Development Environment Beijing China Beihang University State Key Laboratory of Software Development Environment Zhongguancun Laboratory Beijing China

ISBN: (纸本)9798350335231

Visual Grounding is an important part of image annotation generation. The existing methods usually use data alignment based on the similarity calculation of visual text features in location inference and multi-modal fusion, which will lose visual and text information to some extent, and is more likely to make the model overfit the data of specific scenes. To solve this problem, we propose a Visual Grounding Generative Adversarial Network (VGGAN) for visual text fusion using the panoptic transformer. We use the generative adversarial network to generate the prediction, judge the accuracy, and design the visual text transformer according to the panoptic theory. The model can retain the feature information, realize the full interactions between features, thereby better supporting the feature fusion of visual and text. Experimental results on the COCO dataset of complex daily scenes verify the effectiveness of our model, and our model achieves the highest prediction accuracy compared with the state-of-the-art methods. © 2023 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A Secure Mutual Authentication Protocol Based on Visual Cryptography Technique for IoT-Cloud

引用

Chinese Journal of Electronics 2024年第1期33卷 43-57页

作者： Brou Bernard Ehui Chen CHEN Shirui WANG Hua GUO Jianwei LIU School of Cyber Sciences and Technology Beihang University State Key Laboratory of Software Development Environment Beihang University Sino-French Engineering School Beihang University Beihang Hangzhou Innovation Institute Yuhang

Because of the increasing number of threats in the IoT cloud, an advanced security mechanism is needed to guard data against hacking or attacks. A user authentication mechanism is also required to authenticate the user accessing the cloud services. The conventional cryptographic algorithms used to provide security mechanisms in cloud networks are often vulnerable to various cyber-attacks and inefficient against new attacks. Therefore,developing new solutions based on different mechanisms from traditional cryptography methods is required to protect data and users' privacy from attacks. Different from the conventional cryptography method, we suggest a secure mutual authentication protocol based on the visual cryptography technique in this paper. We use visual cryptography to encrypt and decrypt the secret images. The mutual authentication is based on two secret images and *** user requests the ticket from the authentication server(AS) to obtain the permission for accessing the cloud services. Three shared secret keys are used for encrypting and decrypting the authentication process. We analyze the protocol using the Barrows-Abadi-Needham(BAN)-logic method and the results show that the protocol is robust and can protect the user against various attacks. Also, it can provide a secure mutual authentication mechanism.

关键词： Internet of things Visual cryptography Security Mutual authentication IoT-cloud

来源：评论

学校读者我要写书评

暂无评论

Effective Exploration Based on the Structural Information Principles 38

Effective Exploration Based on the Structural Information Pr...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zeng, Xianghua Peng, Hao Li, Angsheng State Key Laboratory of Software Development Environment Beihang University Beijing China Zhongguancun Laboratory Beijing China

Traditional information theory provides a valuable foundation for Reinforcement Learning (RL), particularly through representation learning and entropy maximization for agent exploration. However, existing methods primarily concentrate on modeling the uncertainty associated with RL's random variables, neglecting the inherent structure within the state and action spaces. In this paper, we propose a novel Structural Information principles-based Effective Exploration framework, namely SI2E. Structural mutual information between two variables is defined to address the single-variable limitation in structural information, and an innovative embedding principle is presented to capture dynamics-relevant state-action representations. The SI2E analyzes value differences in the agent's policy between state-action pairs and minimizes structural entropy to derive the hierarchical state-action structure, referred to as the encoding tree. Under this tree structure, value-conditional structural entropy is defined and maximized to design an intrinsic reward mechanism that avoids redundant transitions and promotes enhanced coverage in the state-action space. Theoretical connections are established between SI2E and classical information-theoretic methodologies, highlighting our framework's rationality and advantage. Comprehensive evaluations in the MiniGrid, MetaWorld, and DeepMind Control Suite benchmarks demonstrate that SI2E significantly outperforms state-of-the-art exploration baselines regarding final performance and sample efficiency, with maximum improvements of 37.63% and 60.25%, respectively. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Gererating Twin VFs of Unlabeled and Unpaired VF through Generative Model

Gererating Twin VFs of Unlabeled and Unpaired VF through Gen...

引用

2023 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2023

作者： Zhang, Zhenyu Zhu, Haogang Beihang University State Key Laboratory of Software Development Environment Lab Beijing China

ISBN: (纸本)9798350337488

The accurate analysis and denoising of visual field (VF) measurements play a crucial role in the diagnosis and monitoring of glaucoma, a widespread optic disease leading to visual impairment. This study presents a novel approach that harnesses generative models, specifically the Variational Autoencoder (VAE) and conditional Generative Adversarial Network (cGAN), to address the challenges of denoising VFs posed by the scarcity of labeled and paired VF data. By generating twin VFs whose patterns are consistent with the input of generative models, the denoising network can be trained using these generated twin data under the framework of VF2VF. Thus the applicable range of VF2VF framework expands to encompass unlabeled and unpaired VF datasets. Experiments demonstrate that the denoising network, trained on cGAN-generated data, effectively enhances precision while maintaining accuracy. Furthermore, sensitivity analysis using independent validation datasets showcases the improved sensitivity of the transformed VF vectors to changes. Besides, this study delves into the influence of data volume on denoising performance and compares the performance of VAE and cGAN models. While the method requires extensive generative network training, it unveils promising avenues for advancing VF analysis and denoising by capitalizing on the power of generative models in the absence of labeled and paired data. © 2023 IEEE.

关键词： deterioration detection generative model glaucoma neural network visual field

来源：评论

学校读者我要写书评

暂无评论

Pushing the Limit of Post-Training Quantization

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5556-5570页

作者： Gong, Ruihao Liu, Xianglong Li, Yuhang Fan, Yunqiang Wei, Xiuying Guo, Jinyang Beihang University State Key Laboratory of Complex & Critical Software Environment Beijing100191 China Yale University New HavenCT06511 United States ShanghaiTech University Shanghai201210 China Beihang University State Key Laboratory of Complex & Critical Software Environment Institute of Artificial Intelligence Beijing100191 China

Recently, post-training quantization (PTQ) has become the de facto way to produce efficient low-precision neural networks without long-time retraining. Despite its low cost, current PTQ works fail to succeed under the extremely low-bit setting. In this work, we delve into extremely low-bit quantization and construct a unified theoretical analysis, which provides an in-depth understanding of the reason for the failure of low-bit quantization. According to the theoretical study, we argue that the existing methods fail in low-bit schemes due to significant perturbation on weights and lack of consideration of activation quantization. To this end, we propose Brecq and QDrop to respectively solve these two challenges, based on which a Q-Limit framework is constructed. Then the Q-Limit framework is further extended to support a mixed precision quantization scheme. To the best of our knowledge, this is the first work that can push the limit of PTQ down to INT2. Extensive experiments on various handcrafted and searched neural architectures are conducted for both visual recognition/detection tasks and language processing tasks. Without bells and whistles, our PTQ framework can attain low-bit ResNet and MobileNetV2 comparable with quantization-aware training (QAT), establishing a new state-of-the-art for PTQ. Our code has been open-sourced at https://***/ModelTC/MQBench/. © 1979-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

ISM:intra-class similarity mixing for time series augmentation

引用

Frontiers of Computer Science 2024年第6期18卷 273-275页

作者： Pin LIU Rui WANG Yongqiang HE Yuzhu WANG School of Information Engineering China University of GeosciencesBeijing 100083China State Key Lab of Software Development Environment Beihang UniversityBeijing 100191China

1 *** superior performance of deep models in classification tasks relies heavily on large-scale supervision data with rich features[1].Recent research has shown that improving the feature diversity while expanding the data scale can improve the classification performance[2,3].Time series augmentation possessing the dual strategy is essential in successfully applying deep models in time series classification.

关键词： classification series mixing

来源：评论

学校读者我要写书评

暂无评论

OWL: A LARGE LANGUAGE MODEL FOR IT OPERATIONS 12

OWL: A LARGE LANGUAGE MODEL FOR IT OPERATIONS

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Guo, Hongcheng Yang, Jian Liu, Jiaheng Yang, Liqun Chai, Linzheng Bai, Jiaqi Peng, Junran Hu, Xiaorong Chen, Chao Zhang, Dongfeng Shi, Xu Zheng, Tieqiao Zheng, Liangfan Zhang, Bo Xu, Ke Li, Zhoujun State Key Lab of Software Development Environment Beihang University China Cloudwise Research

With the rapid advancement of IT operations, managing and analyzing large data volumes efficiently for practical applications has become increasingly critical. Natural Language Processing (NLP) techniques have demonstrated remarkable capabilities in various tasks, including named entity recognition, machine translation, and dialogue systems. Recently, Large Language Models (LLMs) have achieved significant improvements across various domain-specific areas. However, there is a noticeable gap in the development of specialized Large Language Models (LLMs) tailored for IT operations. In this paper, we introduce the OWL, a large language model trained on our constructed Owl-Instruct with a wide range of IT-related information. Specifically, limited by the maximum input length, we propose the Homogeneous Markov Context Extension method (HMCE). The mixture-of-adapter strategy is leveraged to improve the parameter-efficient tuning across different domains or tasks. Further, we evaluate the performance of OWL on the Owl-Bench established by us and open IT-related benchmarks. OWL demonstrates superior performance results on IT tasks, which outperforms existing models by significant margins. Moreover, we hope that the findings of our work will provide more insights to revolutionize the techniques of IT operations with specialized LLMs. Our code is available at https://***/HC-Guo/Owl. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Benchmarking

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：