检索结果-内蒙古大学图书馆

40th Conference on Uncertainty in Artificial Intelligence, Uai 2024

作者： Bui, Long Minh Huu, Tho Tran Dinh, Duy Nguyen, Tan Minh Hoang, Trong Nghia FPT Software AI Center Department of Mathematics National University of Singapore Singapore School of Electrical Engineering and Computer Science Washington State University United States

Transformers have increasingly become the de facto method to model sequential data with state-of-the-art performance. Due to its widespread use, being able to estimate and calibrate its modeling uncertainty is important to understand and design robust transformer models. To achieve this, previous works have used Gaussian processes (GPs) to perform uncertainty calibration for the attention units of transformers and attained notable successes. However, such approaches have to confine the transformers to the space of symmetric attention to ensure the necessary symmetric requirement of their GP’s kernel specification, which reduces the representation capacity of the model. To mitigate this restriction, we propose the Correlated Gaussian Process Transformer (CGPT), a new class of transformers whose self-attention units are modeled as cross-covariance between two correlated GPs (CGPs). This allows asymmetries in attention and can enhance the representation capacity of GP-based transformers. We also derive a sparse approximation for CGP to make it scale better. Our empirical studies show that both CGP-based and sparse CGP-based transformers achieve better performance than state-of-the-art GP-based transformers on a variety of benchmark tasks. © 2024 Proceedings of Machine Learning Research. All rights reserved.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

A survey on cross-user federated recommendation

引用

science China(Information sciences) 2025年第4期68卷 7-32页

作者： Enyue YANG Yudi XIONG Wei YUAN Weike PAN Qiang YANG Zhong MING College of Computer Science and Software Engineering Shenzhen University School of Electrical Engineering and Computer Science The University of Queensland WeBank AI Lab WeBank Department of Computer Science and Engineering Hong Kong University of Science and Technology College of Big Data and Internet Shenzhen Technology University Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)

Recommender systems are effective in mitigating information overload, yet the centralized storage of user data raises significant privacy concerns. Cross-user federated recommendation(CUFR) provides a promising distributed paradigm to address these concerns by enabling privacy-preserving recommendations directly on user devices. In this survey, we review and categorize current progress in CUFR, focusing on four key aspects: privacy, security, accuracy, and efficiency. Firstly,we conduct an in-depth privacy analysis, discuss various cases of privacy leakage, and then review recent methods for privacy protection. Secondly, we analyze security concerns and review recent methods for untargeted and targeted *** untargeted attack methods, we categorize them into data poisoning attack methods and parameter poisoning attack methods. For targeted attack methods, we categorize them into user-based methods and item-based methods. Thirdly,we provide an overview of the federated variants of some representative methods, and then review the recent methods for improving accuracy from two categories: data heterogeneity and high-order information. Fourthly, we review recent methods for improving training efficiency from two categories: client sampling and model compression. Finally, we conclude this survey and explore some potential future research topics in CUFR.

关键词： cross-user federated recommendation federated recommendation federated learning recommender systems user privacy

来源：评论

学校读者我要写书评

暂无评论

Experience as Source for Anticipation and Planning: Experiential Policy Learning for Target-driven Recommendation Dialogues

Experience as Source for Anticipation and Planning: Experien...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Dao, Huy Deng, Yang Bui, Khanh-Huyen Le, Dung D. Liao, Lizi Singapore Management University Singapore FPT Software AI Center College of Engineering and Computer Science VinUniversity Viet Nam

ISBN: (纸本)9798891761681

Target-driven recommendation dialogues present unique challenges in dialogue management due to the necessity of anticipating user interactions for successful conversations. Current methods face significant limitations: (I) inadequate capabilities for conversation anticipation, (II) computational inefficiencies due to costly simulations, and (III) neglect of valuable past dialogue experiences. To address these limitations, we propose a new framework, Experiential Policy Learning (EPL), to enhance such dialogues. Specifically, EPL embodies the principle of Learning From Experience, facilitating anticipation with an experiential scoring function that estimates dialogue state potential using similar past interactions stored in long-term memory. Moreover, to demonstrate its flexibility, we introduce Tree-structured EPL (T-EPL) as one possible training-free realization with Large Language Models (LLMs) and Monte-Carlo Tree Search (MCTS). In particular, T-EPL assesses past dialogue states with LLMs while utilizing MCTS to achieve hierarchical and multi-level reasoning. Extensive experiments on two published datasets demonstrate the superiority and efficacy of T-EPL. Code and data are available here. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

A Curious Case of Searching for the Correlation between Training Data and Adversarial Robustness of Transformer Textual Models 62

A Curious Case of Searching for the Correlation between Trai...

引用

Findings of the 62nd Annual Meeting of the Association for Computational Linguistics, ACL 2024

作者： Dang, Cuong Le, Dung D. Le, Thai FPT Software AI Center Viet Nam College of Engineering and Computer Science VinUniversity Viet Nam Department of Computer Science Indiana University United States

ISBN: (纸本)9798891760998

Existing works have shown that fine-tuned textual transformer models achieve state-of-the-art prediction performances but are also vulnerable to adversarial text perturbations. Traditional adversarial evaluation is often done only after fine-tuning the models and ignoring the training data. In this paper, we want to prove that there is also a strong correlation between training data and model robustness. To this end, we extract 13 different features representing a wide range of input fine-tuning corpora properties and use them to predict the adversarial robustness of the fine-tuned models. Focusing mostly on encoder-only transformer models BERT and RoBERTa with additional results for BART, ELECTRA, and GPT2, we provide diverse evidence to support our argument. First, empirical analyses show that (a) extracted features can be used with a lightweight classifier such as Random Forest to predict the attack success rate effectively, and (b) features with the most influence on the model robustness have a clear correlation with the robustness. Second, our framework can be used as a fast and effective additional tool for robustness evaluation since it (a) saves 30x-193x runtime compared to the traditional technique, (b) is transferable across models, (c) can be used under adversarial training, and (d) robust to statistical randomness. Our code is publicly available at https://***/CaptainCuong/RobustText_ACL2024. © 2024 Association for Computational Linguistics.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Revisiting kernel attention with correlated Gaussian process representation 24

Revisiting kernel attention with correlated Gaussian process...

引用

Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence

作者： Long Minh Bui Tho Tran Huu Duy Dinh Tan Minh Nguyen Trong Nghia Hoang FPT Software AI Center and School of Electrical Engineering and Computer Science Washington State University FPT Software AI Center and Department of Mathematics National University of Singapore FPT Software AI Center Department of Mathematics National University of Singapore School of Electrical Engineering and Computer Science Washington State University

Transformers have increasingly become the de facto method to model sequential data with state-of-the-art performance. Due to its widespread use, being able to estimate and calibrate its modeling uncertainty is important to understand and design robust transformer models. To achieve this, previous works have used Gaussian processes (GPs) to perform uncertainty calibration for the attention units of transformers and attained notable successes. However, such approaches have to confine the transformers to the space of symmetric attention to ensure the necessary symmetric requirement of their GP's kernel specification, which reduces the representation capacity of the model. To mitigate this restriction, we propose the Correlated Gaussian Process Transformer (CGPT), a new class of transformers whose self-attention units are modeled as cross-covariance between two correlated GPs (CGPs). This allows asymmetries in attention and can enhance the representation capacity of GP-based transformers. We also derive a sparse approximation for CGP to make it scale better. Our empirical studies show that both CGP-based and sparse CGP-based transformers achieve better performance than state-of-the-art GP-based transformers on a variety of benchmark tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Transaction transmission model for blockchain channels based on non-cooperative games

引用

science China(Information sciences) 2023年第1期66卷 97-114页

作者： Peiyun ZHANG Chenxi LI Mengchu ZHOU Wenjun HUANG Abdullah ABUSORRAH Omaimah O.BAMASAG Engineering Research Center of Digital Forensics of Ministry of Education School of Computer and SoftwareNanjing University of Information Science and Technology School of Computer and Information Anhui Normal University Department of Electrical and Computer Engineering New Jersey Institute of Technology Center of Research Excellence in Renewable Energy and Power Systems Department of Electrical and Computer Engineering Faculty of Engineering and K.A.CARE Energy Research and Innovation CenterKing Abdulaziz University Department of Computer Science Faculty of Computing and Information Technology Center of Excellence in Smart Environment Research King Abdulaziz University

In blockchain networks, transactions can be transmitted through channels. The existing transmission methods depend on their routing information. If a node randomly chooses a channel to transmit a transaction, the transmission may be aborted due to insufficient funds(also called balance) or a low transmission rate. To increase the success rate and reduce transmission delay across all transactions, this work proposes a transaction transmission model for blockchain channels based on non-cooperative game *** balance, channel states, and transmission probability are fully considered. This work then presents an optimized channel transaction transmission algorithm. First, channel balances are analyzed and suitable channels are selected if their balance is sufficient. Second, a Nash equilibrium point is found by using an iterative sub-gradient method and its related channels are then used to transmit transactions. The proposed method is compared with two state-of-the-art approaches: Silent Whispers and Speedy Murmurs. Experimental results show that the proposed method improves transmission success rate, reduces transmission delay,and effectively decreases transmission overhead in comparison with its two competitive peers.

关键词： Blockchain channel transaction transmission channel balance channel state channel transmission probability non-cooperative game

来源：评论

学校读者我要写书评

暂无评论

Mutual Retinex: Combining Transformer and CNN for Image Enhancement

引用

IEEE Transactions on Emerging Topics in Computational Intelligence 2024年第3期8卷 2240-2252页

作者： Jiang, Kui Wang, Qiong An, Zhaoyi Wang, Zheng Zhang, Cong Lin, Chia-Wen Harbin Institute of Technology School of Computer Science and Technology Harbin150001 China Wuhan University National Engineering Research Center for Multimedia Software School of Computer Science Wuhan430072 China Wuhan Polytechnic University School of Mathematics and Computer Science Wuhan430024 China National Tsing Hua University Department of Electrical Engineering Institute of Communications Engineering Hsinchu30013 Taiwan

Images captured in low-light or underwater environments are often accompanied by significant degradation, which can negatively impact the quality and performance of downstream tasks. While convolutional neural networks (CNNs) and Transformer architectures have made significant progress in computer vision tasks, there are few efforts to harmonize them into a more concise framework for enhancing such images. To this end, this study proposes to aggregate the individual capability of self-attention (SA) and CNNs for accurate perturbation removal while preserving background contents. Based on this, we carry forward a Retinex-based framework, dubbed as Mutual Retinex, where a two-branch structure is designed to characterize the specific knowledge of reflectance and illumination components while removing the perturbation. To maximize its potential, Mutual Retinex is equipped with a new mutual learning mechanism, involving an elaborately designed mutual representation module (MRM). In MRM, the complementary information between reflectance and illumination components are encoded and used to refine each other. Through the complementary learning via the mutual representation, the enhanced results generated by our model exhibit superior color consistency and naturalness. Extensive experiments have shown the significant superiority of our mutual learning based method over thirteen competitors on the low-light task and ten methods on the underwater image enhancement task. In particular, our proposed Mutual Retinex respectively surpasses the state-of-the-art method MIRNet-v2 by 0.90 dB and 2.46 dB in PSNR on the LOL 1000 and FIVEK datasets, while with only 19.8% model parameters. © 2017 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Streaming Histogram Publication over Weighted Sliding Windows Under Differential Privacy

引用

Tsinghua science and Technology 2024年第6期29卷 1674-1693页

作者： Xiujun Wang Lei Mo Xiao Zheng Zhe Dang School of Computer Science and Technology Anhui University of TechnologyMa’anshan 243032China Institute for Artificial Intelligence Hefei Comprehensive National Science CenterHefei 230088China Baosight Software(Anhui)Co.Ltd. Ma’anshan 243000China School of Electrical Engineering and Computer Science Washington State UniversityPullman 99164WAUSA

Continuously publishing histograms in data streams is crucial to many real-time applications,as it provides not only critical statistical information,but also reduces privacy leaking *** the importance of elements usually decreases over time in data streams,in this paper we model a data stream by a sequence of weighted sliding windows,and then study how to publish histograms over these windows *** existing literature can hardly solve this problem in a real-time way,because they need to buffer all elements in each sliding window,resulting in high computational overhead and prohibitive storage *** this paper,we overcome this drawback by proposing an online algorithm denoted by Efficient Streaming Histogram Publishing(ESHP)to continuously publish histograms over weighted sliding ***,our method first creates a novel sketching structure,called Approximate-Estimate Sketch(AESketch),to maintain the counting information of each histogram interval at every time instance;then,it creates histograms that satisfy the differential privacy requirement by smartly adding appropriate noise values into the sketching *** experimental results and rigorous theoretical analysis demonstrate that the ESHP method can offer equivalent data utility with significantly lower computational overhead and storage costs when compared to other existing methods.

关键词： differential privacy randomized algorithm streaming data publication weighted sliding window approximate statistics data usability computational complexity

来源：评论

学校读者我要写书评

暂无评论

The Vault: A Comprehensive Multilingual Dataset for Advancing Code Understanding and Generation 3

The Vault: A Comprehensive Multilingual Dataset for Advancin...

引用

2023 Findings of the Association for Computational Linguistics: EMNLP 2023

作者： Manh, Dung Nguyen Le Hai, Nam Dau, Anh T.V. Nguyen, Anh Minh Nghiem, Khanh Guo, Jin Bui, Nghi D.Q. FPT Software AI Center Fulbright University Viet Nam Hanoi University of Science and Technology Viet Nam Viet Nam School of Computer Science McGill University Canada Mila - Quebec AI Institute Canada

ISBN: (纸本)9798891760615

We present The Vault, a dataset of high-quality code-text pairs in multiple programming languages for training large language models to understand and generate code. We present methods for thoroughly extracting samples that use both rule-based and deep learning-based methods to ensure that they contain high-quality pairs of code and text, resulting in a dataset of 43 million high-quality code-text pairs. Our extensive evaluations on common coding tasks including code generation, code search and code summarization show that when fine-tuning Code Large Language Models on The Vault, such models outperform the same models trained on other datasets such as CodeSearchNet. We also provide detailed analyses of our datasets to assess the effects of various programming languages and docstrings on the performance of such models. © 2023 Association for Computational Linguistics.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A Neural Computing-Based Access Control Protocol for ai-Driven Intelligent Flying Vehicles in Industry 5.0-Assisted Consumer Electronics

引用

IEEE Transactions on Consumer Electronics 2024年第1期70卷 3573-3581页

作者： Mahmood, Khalid Tariq, Tayyaba Sangaiah, Arun Kumar Ghaffar, Zahid Saleem, Muhammad Asad Shamshad, Salman National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Douliu64002 Taiwan National Yunlin University of Science and Technology Department of Computer Science and Information Engineering Douliu64002 Taiwan National Yunlin University of Science and Technology International Graduate School of AI Douliu64002 Taiwan Lebanese American University Department of Electrical and Computer Engineering Byblos1102 2801 Lebanon University of Electronic Science and Technology of China School of Computer Science and Engineering Sichuan Chengdu611731 China The University of Lahore Department of Software Engineering Lahore54590 Pakistan

The continuous revolution in Artificial Intelligence (ai) has played a significant role in the development of key consumer applications, including Industry 5.0, autonomous decision-making, fault diagnosis, etc. In practice, autonomous and intelligent robots and flying vehicles (i.e., drones) are also becoming the integral part of Industry 5.0. These vehicles are attaining significant importance nowadays because of their applications in consumer electronics. However, there may be numerous privacy issues in such an infrastructure. Nevertheless, neural computing-driven security solutions have played a significant role in overcoming the security and privacy issues of ai-driven intelligent flying vehicles. Motivated by these facts, this article presents a neural computing-based access control protocol for ai-driven flying vehicles to prevent all possible security threats. A detailed security evaluation of the devised protocol is done formally using the extensively recognized Real-Or-Random (ROR) oracle model. Similarly, the informal security analysis ensures the resistance of our scheme against various potential attacks under the assumptions of CK and DY adversarial models. Moreover, detailed performance scrutiny of the devised protocol with various existing schemes shows the robustness and supremacy of our scheme in terms of computation, communication overhead, and security attributes. © 1975-2011 IEEE.

关键词： Authentication

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：