检索结果-内蒙古大学图书馆

CBPF: A Novel Method For Filtering Poisoned data Based on Composite Backdoor Attacks

IEEE Internet of Things Journal 2025年

作者： Xia, Hanfeng Hong, Haibo Wang, Ruili Sun, Yiru Ding, Hao Zhejiang Gongshang University Zhejiang Key Laboratory of Big Data and Future E-Commerce Technology School of Computer Science and Technology Hangzhou China Massey University School of Mathematical and Computational Sciences Auckland New Zealand

Backdoor attacks involve the injection of a limited quantity of poisoned samples containing triggers into the training dataset. During the inference stage, backdoor attacks can uphold a high level of accuracy for normal examples, yet when presented with trigger-containing instances, the model may erroneously predict them as the targeted class designated by the attacker. This paper addresses the challenge of backdoor attacks by developing a novel method for filtering poisoned samples. We primarily leverage two key characteristics of backdoor attacks: 1) Multiple backdoors can exist simultaneously within a single model;2) The discovery through Composite Backdoor Attack (CBA) that altering two triggers in a sample to new target labels does not compromise the original functionality of the triggers, yet enables the prediction of the data as a new target class when both triggers are present simultaneously. Therefore, a novel three-stage poisoning data filtering approach, known as Composite Backdoor Poisoning Filtering (CBPF), is proposed as an effective solution. Firstly, utilizing the identified distinctions in output between poisoned and clean samples, a subset of data is partitioned to include both poisoned and clean data. Subsequently, benign triggers are incorporated and labels are adjusted to create new target and benign target classes, thereby prompting the poisoned and clean data to be classified as distinct entities during the inference stage. The experimental results indicate that CBPF is successful in filtering out poisoned data produced by seven advanced attacks on CIFAR-10, GTSRB and ImageNet-12. On average, CBPF attains a notable filtering success rate of 99.88% for these attacks on CIFAR-10. Additionally, the model trained on the uncontaminated samples exhibits sustained high accuracy levels. © 2014 IEEE.

关键词： Wiener filtering

来源：评论

学校读者我要写书评

暂无评论

Measuring Diversity of Game Scenarios

arXiv

引用

arXiv 2024年

作者： Li, Yuchen Wang, Ziqi Zhang, Qingquan Yuan, Bo Liu, Jialin Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China Tandon School of Engineering New York University United States School of Data Science Lingnan University Hong Kong

This survey comprehensively reviews the metrics for measuring the diversity of game scenarios, spotlighting the innovative use of procedural content generation and other fields as cornerstones for enriching player experiences through diverse game scenarios. By traversing a wide array of disciplines, from affective modeling and multi-agent systems to psychological studies, our research underscores the importance of diverse game scenarios in gameplay and education. Through a taxonomy of diversity metrics and evaluation methods, we aim to bridge the current gaps in literature and practice, offering insights into effective strategies for measuring and integrating diversity in game scenarios. Our analysis highlights the necessity for a unified taxonomy to aid developers and researchers in crafting more engaging and varied game worlds. This survey not only charts a path for future research in diverse game scenarios but also serves as a handbook for industry practitioners seeking to leverage diversity as a key component of game design and development. Copyright © 2024, The Authors. All rights reserved.

关键词： Taxonomies

来源：评论

学校读者我要写书评

暂无评论

Multi-condition guided depth map completion method based on diffusion model 4

Multi-condition guided depth map completion method based on ...

引用

4th International Conference on computer science and Communication Technology, ICCSCT 2023

作者： Qian, Ren Qiu, Wenfeng Yang, Wenbang Li, Jianhua Wang, Xin’an Zhao, Yong College of Computer Science and Technology Guizhou University Guizhou Guiyang550025 China College of Artificial Intelligence and Big data Luzhou Vocational and Technical College Sichuan Luzhou646000 China School of Electronic and Computer Engineering Shenzhen Graduate School of Peking University Guangdong Shenzhen518055 China Co. Ltd Guangdong Shenzhen518107 China

ISBN: (纸本)9781510671232

The diffusion model has been widely applied in various aspects of artificial intelligence due to its flexible and diverse generative performance. However, there is a lack of research on applying diffusion models in the field of depth map restoration. This is primarily because the diverse generation capabilities are not sufficient to meet the requirements for depth map completion. Depth map completion requires rational completion in the current scene. In order to adapt diffusion models for depth map completion tasks, this paper proposes the Multi Condition Diffusion Model (MCDM). It allows the addition of conditional information to constrain the model’s rational completion of ill-regions. The MultiConditionLN module effectively adds multiple conditions to the depth map completion task. This module uses the completion region mask and the input image as conditions to constrain the model’s generation process. This enables the model to complete the regions that need restoration based on the scene of the input image. The proposed model achieves promising results on depth map datasets. © 2023 SPIE.

关键词： Restoration

来源：评论

学校读者我要写书评

暂无评论

Hyperbolic Geometric Latent Diffusion Model for Graph Generation 41

Hyperbolic Geometric Latent Diffusion Model for Graph Genera...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Fu, Xingcheng Gao, Yisen Wei, Yuecen Sun, Qingyun Peng, Hao Li, Jianxin Li, Xianxian Key Lab of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin China Institute of Artificial Intelligence Beihang University Beijing China School of Software Beihang University Beijing China Beijing Advanced Innovation Center for Big Data and Brain Computing School of Computer Science and Engineering Beihang University Beijing China

Diffusion models have made significant contributions to computer vision, sparking a growing interest in the community recently regarding the application of them to graph generation. Existing discrete graph diffusion models exhibit heightened computational complexity and diminished training efficiency. A preferable and natural way is to directly diffuse the graph within the latent space. However, due to the non-Euclidean structure of graphs is not isotropic in the latent space, the existing latent diffusion models effectively make it difficult to capture and preserve the topological information of graphs. To address the above challenges, we propose a novel geometrically latent diffusion framework HypDiff. Specifically, we first establish a geometrically latent space with interpretability measures based on hyperbolic geometry, to define anisotropic latent diffusion processes for graphs. Then, we propose a geometrically latent diffusion process that is constrained by both radial and angular geometric properties, thereby ensuring the preservation of the original topological properties in the generative graphs. Extensive experimental results demonstrate the superior effectiveness of HypDiff for graph generation with various topologies. Copyright 2024 by the author(s)

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

FullTransNet: Full Transformer with Local-Global Attention for Video Summarization

arXiv

引用

arXiv 2025年

作者： Lan, Libin Jiang, Lu Yu, Tianshu Liu, Xiaojuan He, Zhongshi The College of Computer Science and Engineering Chongqing University of Technology Chongqing400054 China The School of Data Science Chinese University of Hong Kong Shenzhen518172 China The College of Artificial Intelligence Chongqing University of Technology Chongqing401135 China The College of Computer Science Chongqing University Chongqing400044 China

Video summarization mainly aims to produce a compact, short, informative, and representative synopsis of raw videos, which is of great importance for browsing, analyzing, and understanding video content. Dominant video summarization approaches are generally based on recurrent or convolutional neural networks, even recent encoder-only transformers. We propose using full transformer as an alternative architecture to perform video summarization. The full transformer with an encoder-decoder structure, specifically designed for handling sequence transduction problems, is naturally suitable for video summarization tasks. This work considers supervised video summarization and casts it as a sequence-to-sequence learning problem. Our key idea is to directly apply the full transformer to the video summarization task, which is intuitively sound and effective. Also, considering the efficiency problem, we replace full attention with the combination of local and global sparse attention, which enables modeling long-range dependencies while reducing computational costs. Based on this, we propose a transformer-like architecture, named FullTransNet, which has a full encoder-decoder structure with local-global sparse attention for video summarization. Specifically, both the encoder and decoder in FullTransNet are stacked the same way as ones in the vanilla transformer, and the local-global sparse attention is used only at the encoder side. Extensive experiments on two public multimedia benchmark datasets SumMe and TVSum demonstrate that our proposed model can outperform other video summarization approaches, achieving F-Measures of 54.4% on SumMe and 63.9% on TVSum with relatively lower compute and memory requirements, verifying its effectiveness and efficiency. The code and models are publicly available on GitHub. Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

CeRF: Convolutional Neural Radiance Derivative Fields for New View Synthesis

arXiv

引用

arXiv 2023年

作者： Yang, Xiaoyan Lu, Dingbo Liu, Wenjie You, Ling Li, Yang Wang, Changbo School of Computer Science and Technology East China Normal University Shanghai200062 China School of Data Science and Engineering East China Normal University Shanghai200062 China

Recently, Neural Radiance Fields (NeRF) has seen a surge in popularity, driven by its ability to generate high-fidelity novel view synthesized images. However, unexpected"floating ghost" artifacts usually emerge with limited training views and intricate optical phenomena. This issue stems from the inherent ambiguities in radiance fields, rooted in the fundamental volume rendering equation and the unrestricted learning paradigms in multi-layer perceptrons. In this paper, we introduce Convolutional Neural Radiance Fields (CeRF), a novel approach to model the derivatives of radiance along rays and solve the ambiguities through a fully neural rendering pipeline. To this end, a single-surface selection mechanism involving both a modified softmax function and an ideal point is proposed to implement our radiance derivative fields. Furthermore, a structured neural network architecture with 1D convolutional operations is employed to further boost the performance by extracting latent ray representations. Extensive experiments demonstrate the promising results of our proposed model compared with existing state-of-the-art approaches. Copyright © 2023, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

An algorithm for conditional-fault local diagnosis of multiprocessor systems under the MM⁎ model

引用

Theoretical computer science 2024年 987卷

作者： Lv, Yali Lin, Cheng-Kuan Hsu, D. Frank Fan, Jianxi School of Information Technology Henan University of Chinese Medicine Zhengzhou China Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan Undergraduate Degree Program of Systems Engineering and Technology National Yang Ming Chiao Tung University Hsinchu Taiwan Computer Science and Information Engineering Chung Cheng Institute of Technology NDU Taoyuan Taiwan Department of Computer and Information Science Fordham University New York United States School of Computer Science and Technology Soochow University Suzhou China

Diagnosis is a crucial subject for maintaining the reliability of multiprocessor systems. Under the MM⁎ diagnosis model, Sengupta and Dahbura proposed a polynomial-time algorithm with time complexity O(N5) to diagnose a system with N processors. In this paper, we propose a (α,β)-trees combination S(u,X,α,β) and give an algorithm to identify the fault or fault-free status of each processor for conditional local diagnosis under the MM⁎ model. According to our results, a connected network with a (α,β)-trees combination S(u,X,α,β) for a node u is conditionally locally (α+2β−3)-diagnosable at node u and the time complexity of our algorithm to diagnose u is O(α2β+αβ2). As an application, we show that our algorithm can identify the status of each node of n-dimensional star graph Sn if the faulty node number does not exceed 3n−8. Compared with existing algorithms, our algorithm allows more faulty nodes in a multiprocessor system. © 2024 Elsevier B.V.

关键词： Computational complexity

来源：评论

学校读者我要写书评

暂无评论

Blockchain-Based Access Control Architecture for Multi-Domain Environment

SSRN

引用

SSRN 2023年

作者： Li, Yunliang Du, Zhiqiang Fu, Yanfang Ma, Yifan School of Computer Science and Engineering Xi’an Technological University Xi’an710021 China College of Computer and Data Science College of Software Fuzhou University Fuzhou350108 China School of Armament Science and Technology Xi’an Technological University Xi’an710021 China

In multi-domain environment, a plethora of users from diverse domains access information and perform various operations. These users possess intricate permissions, increasing the likelihood of identity falsification, unauthorized access, and privacy breaches during cross-domain interactions. Consequently, implementing an access control architecture is imperative to prevent users from executing illicit activities. This study proposes a novel blockchain-based access control architecture for multi-domain environment. By integrating the multi-domain environment within a federated chain, the architecture employs Decentralized Identifiers (DIDs) for user identification and relies on public/secret key pairs for executing operations. Verifiable credentials are utilized to authorize permissions and release resources, ensuring authenticity verification to deter tampering and forgery. Furthermore, the authorization and access control processes are automated via smart contracts, eliminating the need for human intervention. The proposed architecture enhances security in authentication, permission allocation, and access control discernment. Simulation experiments demonstrate its superior performance concerning runtime, concurrency, and storage efficiency. © 2023, The Authors. All rights reserved.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Fededrated Learning: Research, Challenges, and Future Directions

Fededrated Learning: Research, Challenges, and Future Direct...

引用

Computing, Power and Communication Technologies (IC2PCT), IEEE International Conference on

作者： Kuldeep Singh Kaswan Jagjit Singh Dhatterwal Rishika Yadav Kiran Malik Aanjey Mani Tripathi School of Computing Science and Engineering Galgotias University Greater Noida UP India Artificial Intelligence and Data Science Koneru Lakshmaiah Education Foundation Guntur AP India Department of Computer Science and Engineering Graphic Era Hills University Dehradun Uttarakhand India Department of Computer Science and Engineering MRIEM Rohtak India School of Computing Science and Engineering Galgotias University Greater Noida India

Federated Learning, a disruptive and novel aspect of machine learning, is at the forefront of decentralized, privacy-conscious data processing. This in-depth review study navigates the complex environment of Federated Learning Ecosystems, attempting to extract the core of the topic while shining light on its research horizons, pressing difficulties, and promising approaches. Our trip begins with a look at the fundamental concepts of Federated Learning, emphasizing its importance in the landscape of distributed machine learning paradigms. We track its development via historical milestones and key contributions, offering context for the field’s increasing research. This study focuses on the many features of Federated Learning research, including privacy-preserving techniques, communication protocols, aggregation approaches, and real-world applications. We aggregate major results and advances within these domains, highlighting notable authors and their contributions. However, the route to Federated Learning’s full fulfillment is plagued with difficulties, including questions about privacy, communication efficiency, and scalability. In the future, we will enlighten the growing paths of Federated Learning research, forecasting trends, providing insights into overcoming present problems, and visualizing its integration into diverse fields. To summarize, this review article serves as a guidepost for scholars, practitioners, and enthusiasts wanting to understand the vast domain of Federated Learning. It captures the core of this dynamic discipline, offers strategic counsel for resolving its obstacles, and encourages optimism about its bright future within the domain of distributed machine learning paradigms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Automated Bioacoustic Monitoring for South African Bird Species on Unlabeled data 32

Automated Bioacoustic Monitoring for South African Bird Spec...

引用

32nd International Conference in Central Europe on computer Graphics, Visualization and computer Vision, WSCG 2024

作者： Doell, Michael Kuehn, Dominik Suessle, Vanessa Burnett, Matthew J. Downs, Colleen T. Weinmann, Andreas Hergenroether, Elke Department of Computer Science University of Applied Sciences Darmstadt Schoefferstrasse 3 Darmstadt Germany Centre for Functional Biodiversity School of Life Sciences University of KwaZulu-Natal P/Bag X01 Scottsville Pietermaritzburg3209 South Africa Algorithms for Computer Vision Imaging and Data Analysis Group University of Applied Sciences Darmstadt Schoefferstrasse 3 Darmstadt Germany

Analyses for biodiversity monitoring based on passive acoustic monitoring (PAM) recordings is time-consuming and challenged by the presence of background noise in recordings. Existing models for sound event detection (SED) worked only on certain avian species and the development of further models required labeled data. The developed framework automatically extracted labeled data from available platforms for selected avian species. The labeled data were embedded into recordings, including environmental sounds and noise, and were used to train convolutional recurrent neural network (CRNN) models. The models were evaluated on unprocessed real world data recorded in urban KwaZulu-Natal habitats. The Adapted SED-CRNN model reached a F1 score of 0.73, demonstrating its efficiency under noisy, real-world conditions. The proposed approach to automatically extract labeled data for chosen avian species enables an easy adaption of PAM to other species and habitats for future conservation projects. © 2024 University of West Bohemia. All rights reserved.

关键词： Biotic

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：