检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Hu, Jiacong Gao, Jing Ye, Jingwen Gao, Yang Wang, Xingen Feng, Zunlei Song, Mingli College of Computer Science and Technology Zhejiang University China Robotics Institute School of Computer Science Carnegie Mellon University United States Electrical and Computer Engineering National University of Singapore Singapore School of Software Technology Zhejiang University China State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China Bangsheng Technology Co. Ltd. China

With the rapid development of deep learning, the increasing complexity and scale of parameters make training a new model increasingly resource-intensive. In this paper, we start from the classic convolutional neural network (CNN) and explore a paradigm that does not require training to obtain new models. Similar to the birth of CNN inspired by receptive fields in the biological visual system, we draw inspiration from the information subsystem pathways in the biological visual system and propose Model Disassembling and Assembling (MDA). During model disassembling, we introduce the concept of relative contribution and propose a component locating technique to extract task-aware components from trained CNN classifiers. For model assembling, we present the alignment padding strategy and parameter scaling strategy to construct a new model tailored for a specific task, utilizing the disassembled task-aware components. The entire process is akin to playing with LEGO bricks, enabling arbitrary assembly of new models, and providing a novel perspective for model creation and reuse. Extensive experiments showcase that task-aware components disassembled from CNN classifiers or new models assembled using these components closely match or even surpass the performance of the baseline, demonstrating its promising results for model reuse. Furthermore, MDA exhibits diverse potential applications, with comprehensive experiments exploring model decision route analysis, model compression, knowledge distillation, and more. For more information, please visit https://***/. © 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

A Privacy Protection Scheme for Consumer Electronics data Based on blockchain and AI

引用

IEEE Transactions on Consumer Electronics 2025年

作者： Wang, Luping Yan, Chen Wu, Hongjie Zhu, Fa Kumari, Saru Alenazi, Mohammed J.F State Key Laboratory of Blockchain and Data Security Suzhou University of Science and Technology Suzhou215009 China Zhejiang University Hangzhou310027 China Suzhou University of Science and Technology Suzhou215009 China Nanjing Forestry University College of Information Science College of Artificial Intelligence Nanjing210037 China Chaudhary Charan Singh University Department of Mathematics Uttar Pradesh Meerut250001 India Department of Computer Engineering Riyadh11451 Saudi Arabia

As information technology advances, a significant amount of data related to consumer electronics has emerged, the consumer data is commonly stored on cloud servers, resulting in frequent data leaks and privacy breaches. To tackle these challenges, we propose a novel privacy protection scheme that combines encryption and blockchain technologies. This approach integrates Searchable Encryption, blockchain, and Artificial Intelligence (AI) to facilitate efficient access control and policy enforcement. By employing a decentralized network structure, our scheme minimizes reliance on a central authority, thereby enhancing system security and reliability. User information is independently stored in both blockchain and cloud systems, with digital signatures ensuring data accuracy. To further bolster user privacy, we implement attribute-based encryption for access management and utilize AI for real-time monitoring of potential security threats and suspicious access patterns. Our scheme also features an attribute revocation mechanism to protect sensitive information from unauthorized access. Additionally, customized smart contracts promote equitable interactions between users and cloud service providers. Lastly, the theoretical analyses and experimental results validate that our solution is both secure and efficient in safeguarding consumer data, highlighting its significant innovative contributions to data privacy protection. © 1975-2011 IEEE.

关键词： Sensitive data

来源：评论

学校读者我要写书评

暂无评论

Ethereum Phishing Detection Using Hyperbolic Neural Networks and Temporal Information 22

Ethereum Phishing Detection Using Hyperbolic Neural Networks...

引用

22nd IEEE International Symposium on Parallel and Distributed Processing with Applications, ISPA 2024

作者： Xu, Rong Zhang, Jun Zhu, Xiaoyue Song, Zhaoxiong Yu, Zhi Inner Mongolia University College of Computer Science Hohhot China Zhejiang University The State Key Laboratory of Blockchain and Data Security Hangzhou China Institute of Computing Technology Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798331509712

In recent years, the frequent occurrence of phishing scams on Ethereum has posed serious threats to transaction security and the financial safety of users. This paper proposes an Ethereum phishing scam detection method based on Hyperbolic Neural Networks (HGNNs) and temporal information. The method maps the Ethereum transaction network to hyperbolic space for structural feature extraction, effectively capturing hierarchical structures and complex relationships within the graph, thereby improving the accuracy of phishing scam detection. The model includes a structural feature extraction module and a temporal feature extraction module. It uses HGNN and self-attention mechanism to extract the structural features of the transaction graph, and uses a multi-head attention mechanism to capture the dynamic evolution pattern of the graph. Experimental validation on real Ethereum datasets demonstrates that the proposed model outperforms benchmark models, showcasing its effectiveness. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

A2PO: Towards Effective Offline Reinforcement Learning from an Advantage-aware Perspective

arXiv

引用

arXiv 2024年

作者： Qing, Yunpeng Liu, Shunyu Cong, Jingyuan Chen, Kaixuan Zhou, Yihe Song, Mingli College of Computer Science and Technology Zhejiang University China State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China

Offline reinforcement learning endeavors to leverage offline datasets to craft effective agent policy without online interaction, which imposes proper conservative constraints with the support of behavior policies to tackle the out-of-distribution problem. However, existing works often suffer from the constraint conflict issue when offline datasets are collected from multiple behavior policies, i.e., different behavior policies may exhibit inconsistent actions with distinct returns across the state space. To remedy this issue, recent advantage-weighted methods prioritize samples with high advantage values for agent training while inevitably ignoring the diversity of behavior policy. In this paper, we introduce a novel Advantage-Aware Policy Optimization (A2PO) method to explicitly construct advantage-aware policy constraints for offline learning under mixed-quality datasets. Specifically, A2PO employs a conditional variational auto-encoder to disentangle the action distributions of intertwined behavior policies by modeling the advantage values of all training data as conditional variables. Then the agent can follow such disentangled action distribution constraints to optimize the advantage-aware policy towards high advantage values. Extensive experiments conducted on both the single-quality and mixed-quality datasets of the D4RL benchmark demonstrate that A2PO yields results superior to the counterparts. Our code is available at https://***/Plankson/A2PO. Copyright © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Non-stationary online convex optimization with arbitrary delays 24

Non-stationary online convex optimization with arbitrary del...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Yuanyu Wan Chang Yao Mingli Song Lijun Zhang The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China and School of Software Technology Zhejiang University Ningbo China and Hangzhou High- Tech Zone (Binjiang) Institute of Blockchain and Data Security Hangzhou China The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China and Hangzhou High- Tech Zone (Binjiang) Institute of Blockchain and Data Security Hangzhou China National Key Laboratory for Novel Software Technology Nanjing University Nanjing China and The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China

Online convex optimization (OCO) with arbitrary delays, in which gradients or other information of functions could be arbitrarily delayed, has received increasing attention recently. Different from previous studies that focus on stationary environments, this paper investigates the delayed OCO in non-stationary environments, and aims to minimize the dynamic regret with respect to any sequence of comparators. To this end, we first propose a simple algorithm, namely DOGD, which performs a gradient descent step for each delayed gradient according to their arrival order. Despite its simplicity, our novel analysis shows that the dynamic regret of DOGD can be automatically bounded by $O(\sqrt{\bar{d}T}(P_T+1))$ under mild assumptions, and $O(\sqrt{dT}(P_T+1))$ in the worst case, where $\bar{d}$ and d denote the average and maximum delay respectively, T is the time horizon, and PT is the path-length of comparators. Furthermore, we develop an improved algorithm, which reduces those dynamic regret bounds achieved by DOGD to $O(\sqrt{\bar{d}T(P_T+1)})$ and $O(\sqrt{dT(P_T+1)})$, respectively. The key idea is to run multiple DOGD with different learning rates, and utilize a metaalgorithm to track the best one based on their delayed performance. Finally, we demonstrate that our improved algorithm is optimal in the worst case by deriving a matching lower bound.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Decode the Dark Side of the Language: Applications of LLMs in the Dark Web 9

Decode the Dark Side of the Language: Applications of LLMs i...

引用

9th IEEE International Conference on data Science in Cyberspace, DSC 2024

作者： Chen, Hongfan Diao, Yigang Xiang, Hui Huo, Yixuan Xie, Xiaqing Zhao, Jiapeng Wang, Xuebin Sun, Yanwei Shi, Jinqiao Beijing University of Posts and Telecommunications School of Cyberspace Security Beijing China Zhejiang University The State Key Laboratory of Blockchain and Data Security Hangzhou China Beijing China Data and Technology Support Center Cyberspace Administration of China Beijing China Chinese Academy of Sciences Institute of Information Engineering Beijing China

ISBN: (纸本)9798350391367

Previous studies have discovered that Transformers pre-trained on large scale datasets have a great advantage over traditional methods and can achieve human-level performance in many NLP tasks. Various encoder-only models like BERT and RoBERTa are used in the researches of the language in the Dark Web and most of them have achieved state-of-the-art performance. However, decoder-only models are not frequently used as they were considered to be unsuitable for NLU tasks until recent advancements in large language models. In this paper, we discuss the application of large scale decoder-only models in NLP tasks of real-world Dark Web research. We explore the capabilities of these models with the most realistic datasets in this specific domain. Surprisingly, results show that decoder-only models can achieve similar or better performance compared with state-of-the-art encoder-only model in some cases and show their own unique advantages in such field. We also discuss the limitations of these models and the challenges associated with their application in such field. Our study unveils the potential of decoder-only models in NLP tasks of real-world Dark Web research. © 2024 IEEE.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

PON: Proposal Optimization Network for Temporal Action Proposal Generation 16th

PON: Proposal Optimization Network for Temporal Action Propo...

引用

16th International Conference on Intelligent Computing, ICIC 2020

作者： Peng, Xiaoxiao Du, Jixiang Zhang, Hongbo Department of Computer Science and Technology Huaqiao University Quanzhou China Fujian Key Laboratory of Big Data Intelligence and Security Huaqiao University Quanzhou China Xiamen Key Laboratory of Computer Vision and Pattern Recognition Huaqiao University Quanzhou China

ISBN: (纸本)9783030607951

Temporal action localization is a challenging task in video understanding. Although great progress has been made in temporal action localization, the most advanced methods still have the problem of sharp performance degradation when an action proposal generated. Most methods use sliding windows method or simply group frames according to frame-level scores. These methods are not enough to provide accurate action boundary and maintain reasonable temporal structure. In order to solve these problems, we propose a novel proposal optimization network to generate start score, end score, action score and regression score, and then remove the redundancy by NMS algorithm. In the proposed method, we introduce a metric loss function to maintain the temporal structure of action proposal in the training process. To verify the effectiveness of the proposed method, we have made comparative experiments on ActivityNet-1.3 dataset respectively, and the proposed method has surpassed some of the state-of-the-art methods on the dataset. © 2020, Springer Nature Switzerland AG.

关键词： Passive optical networks

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Convolutional Neural Network with Representation Self-challenge for Fingerprint Liveness Detection

引用

Computers, Materials & Continua 2022年第10期73卷 719-733页

作者： Jie Chen Chengsheng Yuan Chen Cui Zhihua Xia Xingming Sun Thangarajah Akilan School of Computer Science Nanjing University of Information Science and TechnologyNanjing210044China Key Laboratory of Public Security Information Application Based on Big-Data Architecture Ministry of Public SecurityZhejiang Police CollegeHangzhou310053China Jiangsu Yuchi Blockchain Research Institute Nanjing210044China Department of Software Engineering Lakehead UniversityThunder BayON P7B 5E1Canada

Fingerprint identification systems have been widely deployed in many occasions of our daily ***,together with many advantages,they are still vulnerable to the presentation attack(PA)by some counterfeit *** address challenges from PA,fingerprint liveness detection(FLD)technology has been proposed and gradually attracted people’s *** vast majority of the FLD methods directly employ convolutional neural network(CNN),and rarely pay attention to the problem of overparameterization and over-fitting of models,resulting in large calculation force of model deployment and poor model *** at filling this gap,this paper designs a lightweight multi-scale convolutional neural network method,and further proposes a novel hybrid spatial pyramid pooling block to extract abundant features,so that the number of model parameters is greatly reduced,and support multi-scale true/fake fingerprint ***,the representation self-challenge(RSC)method is used to train the model,and the attention mechanism is also adopted for optimization during execution,which alleviates the problem of model over-fitting and enhances generalization of detection ***,experimental results on two publicly benchmarks:LivDet2011 and LivDet2013 sets,show that our method achieves outstanding detection results for blind materials and *** size of the model parameters is only 548 KB,and the average detection error of cross-sensors and cross-materials are 15.22 and 1 respectively,reaching the highest level currently available.

关键词： FLD lightweight multi-scale RSC blind materials

来源：评论

学校读者我要写书评

暂无评论

BSL: Understanding and Improving Softmax Loss for Recommendation 40

BSL: Understanding and Improving Softmax Loss for Recommenda...

引用

40th IEEE International Conference on data Engineering, ICDE 2024

作者： Wu, Junkang Chen, Jiawei Wu, Jiancan Shi, Wentao Zhang, Jizhi Wang, Xiang University of Science and Technology of China MoE Key Laboratory of Brain-inspired Intelligent Perception and Cognition Hefei China Zhejiang Unversity The State Key Laboratory of Blockchain and Data Security China Institute of Artificial Intelligence Institute of Dataspace Hefei Comprehensive National Science Center. China

ISBN: (纸本)9798350317152

Loss functions steer the optimization direction of recommendation models and are critical to model performance, but have received relatively little attention in recent recommendation research. Among various losses, we find Softmax loss (SL) stands out for not only achieving remarkable accuracy but also better robustness and fairness. Nevertheless, the current literature lacks a comprehensive explanation for the efficacy of SL. Toward addressing this research gap, we conduct theoretical analyses on SL and uncover three insights: 1) Optimizing SL is equivalent to performing Distributionally Robust Optimization (DRO) on the negative data, thereby learning against perturbations on the negative distribution and yielding robustness to noisy negatives. 2) Comparing with other loss functions, SL implicitly penalizes the prediction variance, resulting in a smaller gap between predicted values and and thus producing fairer results. Building on these insights, we further propose a novel loss function Bilateral SoftMax Loss (BSL) that extends the advantage of SL to both positive and negative sides. BSL augments SL by applying the same Log-Expectation-Exp structure to positive examples as is used for negatives, making the model robust to the noisy positives as well. Remarkably, BSL is simple and easy-to-implement -requiring just one additional line of code compared to SL. Experiments on four real-world datasets and three representative backbones demonstrate the effectiveness of our proposal. The code is available at https://***/junkangwu/BSL. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

DECRL: A Deep Evolutionary Clustering Jointed Temporal Knowledge Graph Representation Learning Approach

arXiv

引用

arXiv 2024年

作者： Chen, Qian Chen, Ling State Key Laboratory of Blockchain and Data Security College of Computer Science and Technology Zhejiang University China

Temporal Knowledge Graph (TKG) representation learning aims to map temporal evolving entities and relations to embedded representations in a continuous low-dimensional vector space. However, existing approaches cannot capture the temporal evolution of high-order correlations in TKGs. To this end, we propose a Deep Evolutionary Clustering jointed temporal knowledge graph Representation Learning approach (DECRL). Specifically, a deep evolutionary clustering module is proposed to capture the temporal evolution of high-order correlations among entities. Furthermore, a cluster-aware unsupervised alignment mechanism is introduced to ensure the precise one-to-one alignment of soft overlapping clusters across timestamps, thereby maintaining the temporal smoothness of clusters. In addition, an implicit correlation encoder is introduced to capture latent correlations between any pair of clusters under the guidance of a global graph. Extensive experiments on seven real-world datasets demonstrate that DECRL achieves the state-of-the-art performances, outperforming the best baseline by an average of 9.53%, 12.98%, 10.42%, and 14.68% in MRR, Hits@1, Hits@3, and Hits@10, respectively. Copyright © 2024, The Authors. All rights reserved.

关键词： Knowledge graph

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：