检索结果-内蒙古大学图书馆

SoftPatch+: Fully unsupervised anomaly classification and segmentation

Pattern Recognition 2025年 161卷

作者： Wang, Chengjie Jiang, Xi Gao, Bin-Bin Gan, Zhenye Liu, Yong Zheng, Feng Ma, Lizhuang Department of Computer Science and Engineering Shanghai Jiao Tong University China YouTu Lab Tencent China Department of Computer Science and Engineering Southern University of Science and Technology China

Although mainstream unsupervised anomaly detection (AD) (including image-level classification and pixel-level segmentation) algorithms perform well in academic datasets, their performance is limited in practical application due to the ideal experimental setting of clean training data. Training with noisy data is an inevitable problem in real-world anomaly detection but is seldom discussed. This paper is the first to consider fully unsupervised industrial anomaly detection (i.e., unsupervised AD with noisy data). To solve this problem, we proposed memory-based unsupervised AD methods, SoftPatch and SoftPatch+, which efficiently denoise the data at the patch level. Noise discriminators are utilized to generate outlier scores for patch-level noise elimination before coreset construction. The scores are then stored in the memory bank to soften the anomaly detection boundary. Compared with existing methods, SoftPatch maintains a strong modeling ability of normal data and alleviates the overconfidence problem in coreset, and SoftPatch+ has more robust performance which is particularly useful in real-world industrial inspection scenarios with high levels of noise (from 10% to 40%). Comprehensive experiments conducted in diverse noise scenarios demonstrate that both SoftPatch and SoftPatch+ outperform the state-of-the-art AD methods on the MVTecAD, ViSA, and BTAD benchmarks. Furthermore, the performance of SoftPatch and SoftPatch+ is comparable to that of the noise-free methods in conventional unsupervised AD setting. The code of the proposed methods can be found at https://***/TencentYoutuResearch/AnomalyDetection-SoftPatch. © 2024 Elsevier Ltd

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

PExCon: Design and Development of Passive Exoskeleton for Construction 40

PExCon: Design and Development of Passive Exoskeleton for Co...

引用

40th International Symposium on Automation and Robotics in Construction, ISARC 2023

作者： Kansal, Sachin Zhang, Jiansong Computer Science Engineering Department Thapar Institute of Engineering Technology Patiala India Lab School of Construction Management Technology Purdue University Indiana United States

ISBN: (纸本)9780645832204

The concept of teleoperation is crucial in modern and future construction workplaces, as it can be used in remote or challenging/hazardous environments such as workspace in a desert, close to a nuclear reactor, or underwater, among others. The success of teleoperation hinges on its interface, where passive exoskeleton with ergonomics considerations has shown great promise. In the literature, however, the ergonomic design of the passive exoskeleton and the 3-D environment for precise visual feedback is lacking. To address that, the authors target to design and develop an ergonomic inclined and light-weighted passive exoskeleton to support construction teleoperation activities. © ISARC 2023. All rights reserved.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

BEACON: Block-wise Efficient Architecture Co-optimization for DNN-HW-Mapping with Zero-cost Evaluation and Progressive Distillation 37

BEACON: Block-wise Efficient Architecture Co-optimization fo...

引用

37th IEEE International System-on-Chip Conference, SOCC 2024

作者： Yang, Jiun-Kai Chen, Yao-Hua Huang, Chih-Tsun National Tsing Hua University Department of Computer Science Hsinchu Taiwan Industrial Technology Research Institute Electronic and Optoelectronic System Research Lab. Hsinchu Taiwan

ISBN: (纸本)9798350377569

This paper presents the BEACON (block-wise efficient architecture co-optimization) framework, a novel approach to adeptly address the co-optimization challenge of deep neural networks (DNNs), hardware configurations of accelerator architectures, and their mapping strategies. The framework leverages our efficient block-wise zero-cost estimation to accelerate network performance evaluation in a complex macro search space. Additionally, the proposed block-wise progressive distillation enhances the accuracy of co-optimized neural networks. By incorporating an existing hardware-mapping tool, experimental results demonstrate that the BEACON framework reduces the energy-delayproduct by 1.9-22.0 times with a 0.32% accuracy improvement on CIFAR-100, compared with previous optimization works of hardware and its mappings. © 2024 IEEE.

关键词： Cost benefit analysis

来源：评论

学校读者我要写书评

暂无评论

ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool Retrieval 30

ToolRerank: Adaptive and Hierarchy-Aware Reranking for Tool ...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Zheng, Yuanhang Li, Peng Liu, Wei Liu, Yang Luan, Jian Wang, Bin Department of Computer Science and Technology Tsinghua University Beijing China Tsinghua University Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China Xiaomi AI Lab

ISBN: (纸本)9782493814104

Tool learning aims to extend the capabilities of large language models (LLMs) with external tools. A major challenge in tool learning is how to support a large number of tools, including unseen tools. To address this challenge, previous studies have proposed retrieving suitable tools for the LLM based on the user query. However, previously proposed methods do not consider the differences between seen and unseen tools, nor do they take the hierarchy of the tool library into account, which may lead to suboptimal performance for tool retrieval. Therefore, to address the aforementioned issues, we propose ToolRerank, an adaptive and hierarchy-aware reranking method for tool retrieval to further refine the retrieval results. Specifically, our proposed ToolRerank includes Adaptive Truncation, which truncates the retrieval results related to seen and unseen tools at different positions, and Hierarchy-Aware Reranking, which makes retrieval results more concentrated for single-tool queries and more diverse for multi-tool queries. Experimental results show that ToolRerank can improve the quality of the retrieval results, leading to better execution results generated by the LLM. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

SCORE REGULARIZED POLICY OPTIMIZATION THROUGH DIFFUSION BEHAVIOR 12

SCORE REGULARIZED POLICY OPTIMIZATION THROUGH DIFFUSION BEHA...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Chen, Huayu Lu, Cheng Wang, Zhengyi Su, Hang Zhu, Jun Department of Computer Science & Technology Institute for AI BNRist Center Tsinghua-Bosch Joint ML Center THBI Lab Tsinghua University China Guangdong Guangzhou China

Recent developments in offline reinforcement learning have uncovered the immense potential of diffusion modeling, which excels at representing heterogeneous behavior policies. However, sampling from diffusion policies is considerably slow because it necessitates tens to hundreds of iterative inference steps for one action. To address this issue, we propose to extract an efficient deterministic inference policy from critic models and pretrained diffusion behavior models, leveraging the latter to directly regularize the policy gradient with the behavior distribution's score function during optimization. Our method enjoys powerful generative capabilities of diffusion modeling while completely circumventing the computationally intensive and time-consuming diffusion sampling scheme, both during training and evaluation. Extensive results on D4RL tasks show that our method boosts action sampling speed by more than 25 times compared with various leading diffusion-based methods in locomotion tasks, while still maintaining state-of-the-art performance. Code: https://***/thu-ml/SRPO. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

Densely Convolutional BU-NET Framework for Breast Multi-Organ Cancer Nuclei Segmentation through Histopathological Slides and Classification Using Optimized Features

引用

computer Modeling in Engineering & sciences 2024年第12期141卷 2375-2397页

作者： Amjad Rehman Muhammad Mujahid Robertas Damasevicius Faten S.Alamri Tanzila Saba Artificial Intelligence&Data Analytics Lab College of Computer&Information Sciences(CCIS)Prince Sultan UniversityRiyadh11586Saudi Arabia Centre of Real Time Computer Systems Kaunas University of TechnologyKaunasLT-51386Lithuania Department of Mathematical Sciences College of SciencePrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia

This study aims to develop a computational pathology approach that can properly detect and distinguish histology *** is crucial for histopathological image analysis,as it involves segmenting cell ***,challenges exist,such as determining the boundary region of normal and deformed nuclei and identifying small,irregular nuclei *** learning approaches are currently dominant in digital pathology for nucleus recognition and classification,but their complex features limit their practical use in clinical *** existing studies have limited accuracy,significant processing costs,and a lack of resilience and generalizability across diverse *** proposed the densely convolutional Breast U-shaped Network(BU-NET)framework to overcome the mentioned *** study employs BU-NET’s spatial and channel attention methods to enhance segmentation *** inclusion of residual blocks and skip connections in the BU-NEt architecture enhances the process of extracting features and reconstructing the *** enhances the robustness of training and convergence processes by reducing the occurrence of vanishing *** primary objective of BU-NEt is to enhance the model’s capacity to acquire and analyze more intricate features,all the while preserving an efficient working *** BU-NET experiments demonstrate that the framework achieved 88.7%average accuracy,88.8%F1 score for Multi-Organ Nuclei Segmentation Challenge(MoNuSeg),and 91.2%average accuracy,91.8%average F1 for the triple-negative breast cancer(TNBC)*** framework also achieved 93.92 Area under the ROC Curve(AUC)for *** results demonstrated that the technology surpasses existing techniques in terms of accuracy and effectiveness in ***,it showcases the ability to withstand and recover from different tissue types and diseases,indicating possible uses in medical *** research evaluated the efficacy of the proposed method

关键词： Breast cancer histopathology BU-NET deep learning

来源：评论

学校读者我要写书评

暂无评论

Adaptive Attribute-Based Honey Encryption: A Novel Solution for Cloud Data Security

引用

computers, Materials & Continua 2025年第2期82卷 2637-2664页

作者： Reshma Siyal Muhammad Asim Long Jun Mohammed Elaffendi Sundas Iftikhar Rana Alnashwan Samia Allaoua Chelloug School of Computer Science and Engineering Central South UniversityChangsha410083China EIAS LAB College of Computer and Information Sciencesand Center of Excellence in Quantum and Intelligent ComputingPrince Sultan UniversityRiyadh11586Saudi Arabia Gianforte School of Computing Montana State UniversityBozemanMT 59717USA Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh11671Saudi Arabia

A basic procedure for transforming readable data into encoded forms is encryption, which ensures security when the right decryption keys are used. Hadoop is susceptible to possible cyber-attacks because it lacks built-in security measures, even though it can effectively handle and store enormous datasets using the Hadoop Distributed File System (HDFS). The increasing number of data breaches emphasizes how urgently creative encryption techniques are needed in cloud-based big data settings. This paper presents Adaptive Attribute-Based Honey Encryption (AABHE), a state-of-the-art technique that combines honey encryption with Ciphertext-Policy Attribute-Based Encryption (CP-ABE) to provide improved data security. Even if intercepted, AABHE makes sure that sensitive data cannot be accessed by unauthorized parties. With a focus on protecting huge files in HDFS, the suggested approach achieves 98% security robustness and 95% encryption efficiency, outperforming other encryption methods including Ciphertext-Policy Attribute-Based Encryption (CP-ABE), Key-Policy Attribute-Based Encryption (KB-ABE), and Advanced Encryption Standard combined with Attribute-Based Encryption (AES+ABE). By fixing Hadoop’s security flaws, AABHE fortifies its protections against data breaches and enhances Hadoop’s dependability as a platform for processing and storing massive amounts of data.

关键词： Cybersecurity data security cloud storage hadoop encryption and decryption privacy protection attribute-based honey encryption

来源：评论

学校读者我要写书评

暂无评论

Skinned Motion Retargeting with Dense Geometric Interaction Perception 38

Skinned Motion Retargeting with Dense Geometric Interaction ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Ye, Zijie Liu, Jia-Wei Jia, Jia Sun, Shikun Shou, Mike Zheng Department of Computer Science and Technology BNRist Tsinghua University China Key Laboratory of Pervasive Computing Ministry of Education China Show Lab National University of Singapore Singapore

Capturing and maintaining geometric interactions among different body parts is crucial for successful motion retargeting in skinned characters. Existing approaches often overlook body geometries or add a geometry correction stage after skeletal motion retargeting. This results in conflicts between skeleton interaction and geometry correction, leading to issues such as jittery, interpenetration, and contact mismatches. To address these challenges, we introduce a new retargeting framework, MeshRet, which directly models the dense geometric interactions in motion retargeting. Initially, we establish dense mesh correspondences between characters using semantically consistent sensors (SCS), effective across diverse mesh topologies. Subsequently, we develop a novel spatio-temporal representation called the dense mesh interaction (DMI) field. This field, a collection of interacting SCS feature vectors, skillfully captures both contact and non-contact interactions between body geometries. By aligning the DMI field during retargeting, MeshRet not only preserves motion semantics but also prevents self-interpenetration and ensures contact preservation. Extensive experiments on the public Mixamo dataset and our newly-collected ScanRet dataset demonstrate that MeshRet achieves state-of-the-art performance. Code available at https://***/abcyzj/MeshRet. © 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

GUI: A Comprehensive Dataset of Global Urban Infrastructure Based on Geospatial Visual Foundation Models 24

GUI: A Comprehensive Dataset of Global Urban Infrastructure ...

引用

32nd ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, ACM SIGSPATIAL 2024

作者： Han, Zhenyu Zhang, Xin Xi, Yanxin Luo, Yan Xia, Tong Li, Yong Department of Electronic Engineering BNRist Tsinghua University Beijing China Shenzhen International Graduate School Tsinghua University Shenzhen China Department of Computer Science University of Helsinki Helsinki Finland Media Lab Massachusetts Institute of Technology Department of Computing The Hong Kong Polytechnic University Hong Kong Hong Kong Department of Computer Science and Technology University of Cambridge Cambridge United Kingdom

ISBN: (纸本)9798400711077

The substantial social and financial costs of infrastructure identification impede in-depth analyses of sustainable urban design, especially in developing countries. In this paper, we present a novel framework with interactive web visualization based on geospatial visual foundation models. Leveraging this framework, we examine the urban infrastructure information in 1,178 cities worldwide, covering 93, 088 km2 areas. Cross-validation reveals that the overall accuracy of identified infrastructure achieves 67.0%. It sheds light on the sustainable development of cities and exposes the stark inequity in urban infrastructure provision for vulnerable populations. The identified urban infrastructure dataset of this study are available at https://***/tsinghua-fib-lab/GUI, and the interactive web application is at https://***/yz7xbfy3. © 2024 Copyright held by the owner/author(s).

关键词： Artificial Intelligence

来源：评论

学校读者我要写书评

暂无评论

Binary Oriented Feature Selection for Valid Product Derivation in Software Product Line

引用

computers, Materials & Continua 2023年第9期76卷 3653-3670页

作者： Muhammad Fezan Afzal Imran Khan Javed Rashid Mubbashar Saddique Heba G.Mohamed Department of CS&SE International Islamic UniversityIslamabad44000Pakistan Information Technology Services University of OkaraOkara56300Pakistan Departmet of Computer Science MLC LabOkara56300Pakistan Department of Computer Science&Engineering University of Engineering&Technology LahoreNarowal CampusNarowal51601Pakistan Department of Electrical Engineering College of EngineeringPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia

Software Product Line(SPL)is a group of software-intensive systems that share common and variable resources for developing a particular *** feature model is a tree-type structure used to manage SPL’s common and variable features with their different relations and problem of Crosstree Constraints(CTC).CTC problems exist in groups of common and variable features among the sub-tree of feature models more diverse in Internet of Things(IoT)devices because different Internet devices and protocols are ***,managing the CTC problem to achieve valid product configuration in IoT-based SPL is more complex,time-consuming,and ***,the CTC problem needs to be considered in previously proposed approaches such as Commonality VariabilityModeling of Features(COVAMOF)andGenarch+tool;therefore,invalid products are *** research has proposed a novel approach Binary Oriented Feature Selection Crosstree Constraints(BOFS-CTC),to find all possible valid products by selecting the features according to cardinality constraints and cross-tree constraint problems in the featuremodel of ***-CTC removes the invalid products at the early stage of feature selection for the product ***,this research developed the BOFS-CTC algorithm and applied it to,IoT-based feature *** findings of this research are that no relationship constraints and CTC violations occur and drive the valid feature product configurations for the application development by removing the invalid product *** accuracy of BOFS-CTC is measured by the integration sampling technique,where different valid product configurations are compared with the product configurations derived by BOFS-CTC and found 100%*** BOFS-CTC eliminates the testing cost and development effort of invalid SPL products.

关键词： Software product line feature model internet of things crosstree constraints variability management

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：