检索结果-内蒙古大学图书馆

IEEE Joint International information Technology and Artificial Intelligence Conference (ITAIC)

作者： Dan Wang Qinglong Meng Huijuan Hao Guanghe Cheng Yongwei Tang Jinqiang Bai Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

In this paper, we address the issue of gearbox fault diagnosis under changing working conditions and the limitations of migration learning algorithms that are only applicable to scenarios with identical labels in both the source and target domains. Our proposed solution is a weighted adversarial migration network model that is based on weighted adversarial migration. To transfer knowledge from the source domain to the target domain, we use transfer learning. Additionally, to tackle the problem of new fault classes in the target domain, we utilize a deep residual network to extract similar data features between different working conditions. This allows us to train a labelled classifier to identify known fault classes and some new target samples in both the source and target domains. We generate weights using auxiliary classifiers and discriminators to assess the importance of matching feature distributions between the known and unknown classes of samples. The experimental results demonstrate that our model can successfully migrate labelled datasets to unlabelled similar datasets in open set scenarios. It also has excellent fault diagnosis capabilities and high accuracy for gearboxes with variable operating conditions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Job2Vec: A Self-Supervised Contrastive Learning Based HPC Job Power Consumption Prediction Framework

Job2Vec: A Self-Supervised Contrastive Learning Based HPC Jo...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Jie Zhang Jian Song Xiang Li Xuesen Tian Zhigang Zhao Lu Wu Chunxiao Wang Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

In the context of the rapid development of big data and artificial intelligence, the field of HPC is facing significant challenges in energy consumption. To address this challenge, this paper proposes a power consumption prediction framework for HPC jobs based on self-supervised contrastive learning, named Job2Vec. Firstly, HPC jobs are clustered into different clusters based on job logs and power consumption curve shapes. Next, high-quality features of power consumption data at different granularities within each cluster are captured through self-supervised contrastive learning. Lastly, an incremental update strategy is introduced to handle the dynamic changes in HPC data. The experimental results demonstrate that, on the Jinan Supercomputing dataset comprising 1170 HPC jobs, the Job2Vec framework outperforms the baseline models overall in terms of prediction performance. Additionally, we also investigated the model’s performance in single-step and multi-step forecasting, the impact of different indicators on prediction accuracy, and conducted ablation experiments to confirm the indispensability of each module. This is the first time that self-supervised contrastive learning has been introduced into the HPC field, aiming to provide support for energy scheduling in data centers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inserting anybody in diffusion models via celeb basis 23

Inserting anybody in diffusion models via celeb basis

引用

Proceedings of the 37th International Conference on Neural information Processing Systems

作者： Ge Yuan Xiaodong Cun Yong Zhang Maomao Li Chenyang Qi Xintao Wang Ying Shan Huicheng Zheng School of Computer Science and Engineering Sun Yat-sen University and Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China and Guangdong Key Laboratory of Information Security Technology and Tencent AI Lab Tencent AI Lab The Hong Kong University of Science and Technology School of Computer Science and Engineering Sun Yat-sen University and Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China and Guangdong Key Laboratory of Information Security Technology

Exquisite demand exists for customizing the pretrained large text-to-image model, e.g. Stable Diffusion, to generate innovative concepts, such as the users themselves. However, the newly-added concept from previous customization methods often shows weaker combination abilities than the original ones even given several images during training. We thus propose a new personalization method that allows for the seamless integration of a unique individual into the pre-trained diffusion model using just one facial photograph and only 1024 learnable parameters under 3 minutes. So we can effortlessly generate stunning images of this person in any pose or position, interacting with anyone and doing anything imaginable from text prompts. To achieve this, we first analyze and build a well-defined celeb basis from the embedding space of the pre-trained large text encoder. Then, given one facial photo as the target identity, we generate its own embedding by optimizing the weight of this basis and locking all other parameters. Empowered by the proposed celeb basis, the new identity in our customized model showcases a better concept combination ability than previous personalization methods. Besides, our model can also learn several new identities at once and interact with each other where the previous customization model fails to. Project page is at: http://***. Code is at: https://***/ygtxr1997/CelebBasis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Planning with Learned Dynamic Model for Unsupervised Point Cloud Registration 30

Planning with Learned Dynamic Model for Unsupervised Point C...

引用

30th International Joint Conference on Artificial Intelligence, IJCAI 2021

作者： Jiang, Haobo Qian, Jianjun Xie, Jin Yang, Jian PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9780999241196

Point cloud registration is a fundamental problem in 3D computer vision. In this paper, we cast point cloud registration into a planning problem in reinforcement learning, which can seek the transformation between the source and target point clouds through trial and error. By modeling the point cloud registration process as a Markov decision process (MDP), we develop a latent dynamic model of point clouds, consisting of a transformation network and evaluation network. The transformation network aims to predict the new transformed feature of the point cloud after performing a rigid transformation (i.e., action) on it while the evaluation network aims to predict the alignment precision between the transformed source point cloud and target point cloud as the reward signal. Once the dynamic model of the point cloud is trained, we employ the cross-entropy method (CEM) to iteratively update the planning policy by maximizing the rewards in the point cloud registration process. Thus, the optimal policy, i.e., the transformation between the source and target point clouds, can be obtained via gradually narrowing the search space of the transformation. Experimental results on ModelNet40 and 7Scene benchmark datasets demonstrate that our method can yield good registration performance in an unsupervised manner. © 2021 International Joint Conferences on Artificial Intelligence. All rights reserved.

关键词： Dynamic models

来源：评论

学校读者我要写书评

暂无评论

Cluster Caching Strategy Based on User Characteristics in Edge networks

Cluster Caching Strategy Based on User Characteristics in Ed...

引用

Asia-Pacific network Operations and Management Symposium (APNOMS)

作者： Ye Zhang Wei Zhang Hao Hao Kai Zhang Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Key Laboratory of Computing Power Network and Information Security Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China

To meet the growing traffic demand and relieve the pressure on the core network, edge caching and device-to-device (D2D) communication are used in next generation mobile networks. In this paper, based on a three-way collaboration between user devices (UEs), base stations (BSs) and cloud data center, caching the content most likely to be requested can facilitate direct content delivery without the need to fetch content from remote cloud data center in the mobile edge networks. This reduces the average retrieval latency of content, reduces the backhaul burden and minimises duplicate content delivery. Specifically, we first use a spectral clustering algorithm to cluster users based on their social and physical attributes, and then associate each cluster with a different base station in order to efficiently cache popular contents. Finally, we propose a new caching policy based on content cache value and user activity to take full account of the metrics that influence caching decisions. Experimental results show that our proposed policy can effectively reduce content access latency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Boosting Adverse Weather Crowd Counting via Multi-queue Contrastive Learning

arXiv

引用

arXiv 2024年

作者： Pan, Tianhang Jia, Xiuyi School of Computer Science and Engineering Nanjing University of Science and Technology Jiangsu Nanjing210094 China Guangxi Key Lab of Multi-source Information Mining & Security Guangxi Normal University Guangxi Guilin541004 China

Currently, most crowd counting methods have outstanding performance under normal weather conditions. However, they often struggle to maintain their performance in extreme and adverse weather conditions due to significant differences in the domain and a lack of adverse weather images for training. To address this issue and enhance the model’s robustness in adverse weather, we propose a two-stage crowd counting method. Specifically, in the first stage, we introduce a multi-queue MoCo contrastive learning strategy to tackle the problem of weather class imbalance. This strategy facilitates the learning of weather-aware representations by the model. In the second stage, we propose to refine the representations under the guidance of contrastive learning, enabling the conversion of the weather-aware representations to the normal weather domain. While significantly improving the robustness, our method only marginally increases the weight of the model. In addition, we also create a new synthetic adverse weather dataset. Extensive experimental results show that our method achieves competitive performance. © 2024, CC BY.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Image-Text Retrieval with Intra-Modal Soft labels

SSRN

引用

SSRN 2022年

作者： Liu, Haoliang Zhang, Yu School of Computer Science and Engineering The Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China

Image-text retrieval seeks to find relevant images from a huge image corpus given a textual description, and vice versa. The critical challenge of this task is to well learn image-text alignments to measure the image-text similarity. Existing works normally label the negative/positive image-text pair with a binary label. They assume that a text is only relevant to the paired images and irrelevant to other images in the corpus. However, using the binary labels during training may be sub-optimal for several reasons. First, a binary label can not reveal the precise similarity between an image and a text. Second, the annotated labels may contain noise. In this paper, we study the idea of training the model with the soft label information in which each image-text pair is associated with a specific soft label. Intuitively, if the textual description of an image shares a high textual similarity with a text, the image should also be highly relevant to the text. Therefore, we determine the soft label between the cross-modal (image, text) pair by the similarity between the intra-modal (image caption, text) pair, which is termed as intra-modal soft label (ISL). In the experiment, we evaluated our method on benchmark datasets, which demonstrate the effectiveness of the proposed method. © 2022, The Authors. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

D4A: An efficient and effective defense across agnostic adversarial attacks

引用

Neural networks 2025年 183卷 106938页

作者： Li, Xianxian Gan, Zeming Bai, Yan Su, Linlin Li, De Wang, Jinyan Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin China Guangxi Key Lab of Multi-source Information Mining and Security Guangxi Normal University Guilin China School of Computer Science and Engineering Guangxi Normal University Guilin China School of Engineering and Technology University of Washington Tacoma TacomaWA United States

Recent studies show that Graph Neural networks (GNNs) are vulnerable to structure adversarial attacks, which draws attention to adversarial defenses in graph data. Previous defenses designed heuristic defense strategies for specific attacks or graph properties, and are no longer sufficiently robust across all these attacks. To address this problem, we discuss the abnormal behaviors of GNNs in structure perturbations from a posterior distribution perspective. We suggest that the structural vulnerability of GNNs stems from their dependence on local graph smoothing, which can also lead to unfitting — a first-found phenomenon specific to the graph domain. We demonstrate that abnormal behaviors, except for unfitting, can attribute to a posterior distribution shift. To intrinsically prevent the occurrence of abnormal behaviors, we first propose smooth-less message passing to enhance the tolerance with respect to structure perturbations, while significantly mitigating the unfitting. We also propose the distribution shift constraint to restrict other abnormal behaviors of our model. Our approach is evaluated on six different datasets across over four kinds of attacks and compared to 11 representative baselines. The experimental results show that our method improves the defense performance across various attacks, and provides a great trade-off between accuracy and adversarial robustness. © 2024 Elsevier Ltd

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

UAV-RIS-Aided Energy-Efficient and QoS-Aware Emergency Communications Based on DRL 100

UAV-RIS-Aided Energy-Efficient and QoS-Aware Emergency Commu...

引用

100th IEEE Vehicular Technology Conference, VTC 2024-Fall

作者： Hu, Jiawei Ju, Ying Wang, Haoyu Liu, Lei Pei, Qingqi Shee, Y.G. Zhu, Xiaojie Wu, Celimuge University of California Department of Electrical Engineering and Computer Science IrvineCA92697 United States Xidian University Guangzhou Institute of Technology Guangzhou510555 China Xi'an University of Posts Telecommunications Shaanxi Key Laboratory of Information Communication Network and Security Xi'an710121 China Universiti Tunku Abdul Rahman Department of Electrical and Electronic Engineering Kajang43000 Malaysia King Abdullah University of Science and Technology Department of Computer Electrical and Mathematical Sciences and Engineering Thuwal239556900 Saudi Arabia The University of Electro-Communications Graduate School of Informatics and Engineering Tokyo182-8585 Japan

ISBN: (纸本)9798331517786

Ensuring reliable communication can be incredibly challenging in emergencies due to the breakdown of conventional infrastructure. However, a promising solution is on the horizon: the integration of reconfigurable intelligent surfaces (RIS) onto unmanned aerial vehicles (UAV), known as UAV-RIS. This innovative approach holds the potential to offer agile and adaptable communication services during crises, overcoming the limitations of traditional systems. This paper establishes an innovative UAV-RIS system with an active RIS to enhance the uplink communication between ground devices (GDs) and the air base station (ABS). We present an advanced communication strategy utilizing deep reinforcement learning (DRL) for UAV-RIS-supported uplink communication in dynamic emergencies. This scheme is designed to optimize the energy efficiency of the UAV-RIS communication system while adhering to quality of service (QoS) constraints for all GDs. It achieves this by jointly optimizing the trajectory of the UAV-RIS and the phase of the active RIS, ensuring efficient and reliable communication in challenging environments. To optimize the performance of the system, we propose a hierarchical Proximal Policy Optimization (H-PPO) algorithm and the upper and lower layers of H-PPO optimize the trajectory and phase control, respectively. Simulation results demonstrate that our scheme can effectively learn the communication strategy to enhance the performance of dynamic emergency communication networks. © 2024 IEEE.

关键词： active reconfigurable intelligent surface deep reinforcement learning quality of service UAV trajectory

来源：评论

学校读者我要写书评

暂无评论

Image-Text Matching with Intra-Modal Knowledge

SSRN

引用

SSRN 2022年

作者： Liu, Haoliang Zhang, Yu School of Computer Science and Engineering The Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China

In this paper, we study the problem of image-text matching. Image-text matching aims to train an model that learns to identify whether or not an (image, caption) input pair constitutes an aligned pair. To boost image-text matching, the soft label is introduced by knowledge distillation. Its benefit is generally attributed to the privileged information on similarities among image-text pairs provided by the teacher model. However, training a stronger teacher is computationally heavy and inefficient. In this paper, we propose a novel parameter-free Soft labels Generation(SLG) method for image-text matching which can produce soft labels with privileged information without the need of training a heavy teacher model. We propose to use intra-modal knowledge to generate the soft label for each image-text pair, which provides an alternative to knowledge distillation. Our method is especially applicable to cases where a stronger teacher model is not available, or only limited computation resources are provided. Systematic experiments conducted on public benchmarks demonstrate that our SLG is empirically effective. Our SLG achieves about 2%-3% recall@1 boosting than the baseline. © 2022, The Authors. All rights reserved.

关键词： Personnel training

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：