检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Guo, Wenzhong Wang, Hanling Ke, Xiao Fujian Provincial Key Laboratory of Networking Computing and Intelligent Information Processing College of Computer and Data Science Fuzhou University Fujian Fuzhou350116 China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou University Fuzhou350116 China

Stereoscopic images usually consist of a left view, a right view, and depth information. Stereoscopic/3D image quality assessment (SIQA) is often more challenging compared to 2D-IQA due to the usual scene differences between left and right views and the complexity of the binocular vision fusion process. To address the problem of quality prediction bias of multi-distortion images, we investigated the visual physiology and the processing of visual information by the primary visual cortex of the human brain and proposed a no-reference stereoscopic image quality evaluation method. The method mainly includes an innovative end-to-end NR-SIQA neural network with a picture patch generation algorithm. The algorithm generates a saliency map by fusing the left and right views and then guides the image cropping in the database based on the saliency map. The proposed models are validated and compared based on publicly available databases. The results show that the model and algorithm together outperform the state-of-the-art NR-SIQA metric in the LIVE 3D database and the WIVC 3D database, and have excellent results in the specific noise metric. The model generalization experiments demonstrate a certain degree of generality of our proposed model. © 2024, The Authors. All rights reserved.

关键词： Database systems

来源：评论

学校读者我要写书评

暂无评论

LEARNING MONOTONIC ATTENTION IN TRANSDUCER FOR STREAMING GENERATION

arXiv

引用

arXiv 2024年

作者： Ma, Zhengrui Feng, Yang Zhang, Min Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Streaming generation models are increasingly utilized across various fields, with the Transducer architecture being particularly popular in industrial applications. However, its input-synchronous decoding mechanism presents challenges in tasks requiring non-monotonic alignments, such as simultaneous translation, leading to suboptimal performance in these contexts. In this research, we address this issue by tightly integrating Transducer’s decoding with the history of input stream via a learnable monotonic attention mechanism. Our approach leverages the forward-backward algorithm to infer the posterior probability of alignments between the predictor states and input timestamps, which is then used to estimate the context representations of monotonic attention in training. This allows Transducer models to adaptively adjust the scope of attention based on their predictions, avoiding the need to enumerate the exponentially large alignment space. Extensive experiments demonstrate that our MonoAttn-Transducer significantly enhances the handling of non-monotonic alignments in streaming generation, offering a robust solution for Transducer-based frameworks to tackle more complex streaming generation tasks. Codes: https://***/ictnlp/MonoAttn-Transducer. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Multivariate Time Series Anomaly Detection by Feature Decoupling in Federated Learning Scenarios

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年

作者： He, Yifan Ding, Xi Tang, Yateng Guan, Jihong Zhou, Shuigeng Fudan University School of Computer Science Shanghai200433 China Shanghai Key Laboratory of Intelligent Information Processing Shanghai200433 China Tencent Weixin Group Guangzhou510308 China Tongji University Department of Computer Science and Technology Shanghai201804 China Ministry of Education Key Laboratory of Embedded System and Service Computing Tongji University Shanghai200092 China

Anomalies are usually regarded as data errors or novel patterns previously unseen, which are quite different from most observed data. Accurate detection of anomalies is crucial in various application scenarios. This paper focuses on unsupervised anomaly detection from Multivariate Time Series (MTS), as real-world data collected from sources such as wearable devices, medical equipment and industrial machines typically manifest as MTS and are often unlabeled. Anomaly detection in MTS represents a data-driven challenge that traditionally requires substantial centralized data for training models. However, in practice, data are frequently distributed among multiple institutions, with privacy concerns restricting unrestricted access. To address these issues, we introduce Feature Decoupling Federated Learning (FDFL), an approach designed to collaboratively train a representation learning network over multiple clients for unsupervised anomaly detection in MTS. Unlike previous methods that simply integrate MTS anomaly detection algorithms with federated learning strategies, FDFL specifically addresses heterogeneity among clients by decoupling the representation network into shared and private branches through a contrastive learning mechanism. This method aggregates shared parameters during each federated round while maintaining client-specific private parameters locally. Additionally, we develop a self-attention block that integrates the representations derived from both shared and private parameters to reconstruct MTS and identify anomalies based on reconstruction errors. Extensive experiments conducted on three publicly available datasets demonstrate that FDFL outperforms existing algorithms in most cases, highlighting the effectiveness and superiority of our proposed method in MTS anomaly detection. © 2020 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A Non-autoregressive Generation Framework for End-to-End Simultaneous Speech-to-Speech Translation

arXiv

引用

arXiv 2024年

作者： Ma, Zhengrui Fang, Qingkai Zhang, Shaolei Guo, Shoutao Feng, Yang Zhang, Min Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Simultaneous translation models play a crucial role in facilitating communication. However, existing research primarily focuses on text-to-text or speech-to-text models, necessitating additional cascade components to achieve speech-to-speech translation. These pipeline methods suffer from error propagation and accumulate delays in each cascade component, resulting in reduced synchronization between the speaker and listener. To overcome these challenges, we propose a novel non-autoregressive generation framework for simultaneous speech translation (NAST-S2x1), which integrates speech-to-text and speech-to-speech tasks into a unified end-to-end framework. We develop a non-autoregressive decoder capable of concurrently generating multiple text or acoustic unit tokens upon receiving fixed-length speech chunks. The decoder can generate blank or repeated tokens and employ CTC decoding to dynamically adjust its latency. Experimental results show that NAST-S2x outperforms state-of-the-art models in both speech-to-text and speech-to-speech tasks. It achieves high-quality simultaneous interpretation within a delay of less than 3 seconds and provides a 28× decoding speedup in offline generation. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Large Model-Based Data Augmentation for Imbalanced Text Classification

Large Model-Based Data Augmentation for Imbalanced Text Clas...

引用

Artificial Intelligence, Networking and information Technology (AINIT), International Seminar on

作者： Dawei Zhang Rongxin Mi Peiyao Zhou Dawei Jin Manman Zhang Tianhang Song School of Computer Science Jiangsu University of Science and Technology Zhenjiang Jiangsu China National Computer Network Emergency Response Technical Team/Coordination Center of China Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences (CAS) Beijing China

ISBN: (数字)9798350385557

ISBN: (纸本)9798350385564

This study focuses on the application of large models to deal with imbalanced data problems in text classification. In view of the central position of text in web data and the negative impact of class imbalance on classifier performance, researchers have explored the method of using large models to generate high-quality minority class samples to enhance model performance. This paper reviews the technical progress of machine learning, deep learning, and large language models and their applications in text classification tasks. Although large models perform well in complex tasks due to their excellent language understanding ability, traditional machine learning and deep learning methods are popular in text classification scenarios that require fast response due to their simple structure and higher computational efficiency. This study proposes a data augmentation technique inspired by SMOTE, which uses a large language model combined with a simple prompt engineering strategy to generate high-quality minority samples. The experimental results show that the proposed method significantly improves the macro average precision, recall and F1 score on multiple text classification models, and effectively alleviates the challenge of class imbalance.

关键词： Deep learning Seminars Computational modeling Large language models Text categorization Data augmentation Data models

来源：评论

学校读者我要写书评

暂无评论

Combining State Detection with Knowledge Transfer for Constrained Multi-objective Optimization

Combining State Detection with Knowledge Transfer for Constr...

引用

International Conference on Tools for Artificial Intelligence (ICTAI)

作者： Jinhua Zheng Kaixi Yang Juan Zou Shengxiang Yang Key Laboratory of Intelligent Computing and Information Processing Xiangtan University China School of Computer Science and Informatics De Montfort University U.K

The main challenge in studying constrained multi-objective optimization problems (CMOPs) is reasonably balancing convergence, diversity, and feasibility. One of the most successful solutions to this challenge is the co-evolutionary frame-work, an algorithm in which multiple populations cooperate and complement each other, with different but interdependent populations addressing different but related problems. However, the effectiveness of existing algorithms for information exchange between various populations is not apparent. This paper proposes a new algorithm named SDKT using population state detection and knowledge transfer. The method has dual stages (i.e., knowledge acquisition and knowledge reception) and dual populations. Specifically, by restarting the strategy, these two populations (i.e., mainPop and auxPop) first explore more feasible regions with and without constraints. Then, in the knowledge receiving stage, ma i nPop and auxPop provide effective information to promote each other's approach to constrained PF (CPF) and unconstrained PF (UPF), respectively. Extensive experiments on three well-known test suites and three real-world problem studies fully demonstrate that SDKT is more competitive than five state-of-the-art constrained multi-objective evolutionary algorithms.

关键词： Knowledge acquisition Sociology Diversity reception Evolutionary computation Statistics Artificial intelligence Optimization

来源：评论

学校读者我要写书评

暂无评论

Hamiltonian Boundary Value Methods (HBVMs) for functional differential equations with piecewise continuous arguments

arXiv

引用

arXiv 2024年

作者： Gurioli, Gianmarco Wang, Weijie Yan, Xiaoqiang Università degli Studi di Firenze Florence Italy School of Mathematics and Computational Science Hunan Key Laboratory for Computation and Simulation in Science and Engineering Xiangtan University Hunan411105 China Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education Xiangtan University Hunan411105 China

In this paper, a class of high-order methods to numerically solve Functional Differential Equations with Piecewise Continuous Arguments (FDEPCAs) is discussed. The framework stems from the expansion of the vector field associated with the reference differential equation along the shifted and scaled Legendre polynomial orthonormal basis, working on a suitable extension of Hamiltonian Boundary Value Methods. Within the design of the methods, a proper generalization of the perturbation results coming from the field of ordinary differential equations is considered, with the aim of handling the case of FDEPCAs. The error analysis of the devised family of methods is performed, while a few numerical tests on Hamiltonian FDEPCAs are provided, to give evidence to the theoretical findings and show the effectiveness of the obtained resolution *** Codes 65L05, 65L03, 65L06, 65P10 Copyright © 2024, The Authors. All rights reserved.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

Dual-axial self-attention network for text classification

引用

science China(information sciences) 2021年第12期64卷 80-90页

作者： Xiaochuan ZHANG Xipeng QIU Jianmin PANG Fudong LIU Xingwei LI State Key Laboratory of Mathematical Engineering and Advanced Computing Shanghai Key Laboratory of Intelligent Information Processing Fudan University School of Computer Science Fudan University

Text classification is an important task in natural language processing and numerous studies aim to improve the accuracy and efficiency of text classification models. In this study, we propose an effective and efficient text classification model which is based on self-attention solely. The recently proposed multidimensional self-attention significantly improved the performance of self-attention. However, existing models suffer from two major limitations:(1) the previous multi-dimensional self-attention models are quite timeconsuming;(2) the dependencies of elements along the feature axis are not taken into account. To overcome these problems, in this paper, a much more computational efficient multi-dimensional self-attention model is proposed, and two parallel self-attention modules, called dual-axial self-attention, are applied to capture rich dependencies along the feature axis as well as the text axis. A text classification model is then *** experimental results on eight representative datasets show that the proposed text classification model can obtain state-of-the-art results and the proposed self-attention outperforms conventional self-attention models.

关键词： text classification dual-axial self-attention feature-axial dependency

来源：评论

学校读者我要写书评

暂无评论

Partial Relaxed Optimal Transport for Denoised Recommendation

Partial Relaxed Optimal Transport for Denoised Recommendatio...

引用

2022 Workshop on Deep Learning for Search and Recommendation, DL4SR 2022

作者： Tan, Yanchao Yang, Carl Wei, Xiangyu Wu, Ziyue Liu, Weiming Zheng, Xiaolin The College of Computer and Data Science Fuzhou University Fuzhou350116 China Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou350116 China The Department of Computer Science Emory University Atlanta30322 United States The College of Computer Science Zhejiang University Hangzhou310027 China The School of Management Zhejiang University Hangzhou310058 China

The interaction data used by recommender systems (RSs) inevitably include noises resulting from mistaken or exploratory clicks, especially under implicit feedbacks. Without proper denoising, RS models cannot effectively capture users’ intrinsic preferences and the true interactions between users and items. To address such noises, existing methods mostly rely on auxiliary data which are not always available. In this work, we ground on Optimal Transport (OT) to globally match a user embedding space and an item embedding space, allowing both non-deep and deep RS models to discriminate intrinsic and noisy interactions without supervision. Specifically, we firstly leverage the OT framework via Sinkhorn distance to compute the continuous many-to-many user-item matching scores. Then, we relax the regularization in Sinkhorn distance to achieve a closed-form solution with a reduced time complexity. Finally, to consider individual user behaviors for denoising, we develop a partial OT framework to adaptively relabel user-item interactions through a personalized thresholding mechanism. Extensive experiments show that our framework can significantly boost the performances of existing RS models. © 2022 Copyright for this paper by its authors.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Lossless and Universal 3D Object Encryption with Differentiated Visual Effects upon Decryption: A Novel Paradigm

引用

IEEE Transactions on Dependable and Secure computing 2025年

作者： Zhao, Ruoyu Zhang, Yushu Mou, Jun Puech, William Weng, Jian School of Computing and Artificial Intelligence Jiangxi University of Finance and Economics Nanchang330013 China Jiangxi Provincial Key Laboratory of Multimedia Intelligent Processing Nanchang330013 China School of Information Science and Engineering Dalian Polytechnic University Dalian116000 China Centre National de la Recherche Scientifique Université de Montpellier Montpellier34095 France College of Information Science and Technology Jinan University Guangzhou510632 China

A recently proposed 3D object encryption scheme enables hierarchical decryption, allowing a single encryption to display varied visual effects upon decryption. This has potential applications in complex scenarios where 3D objects require access at different security levels. However, it has two main issues: lossless decryption is not possible due to precision loss from the IEEE 754 standard, and it is customized for a specific AES algorithm, limiting support for others. Motivated by this, a novel paradigm for 3D object encryption with differentiated visual effects upon decryption is proposed. In this paradigm, the precision loss is preserved within the 3D encrypted object, allowing the 3D object after fully decrypting to be identical to the original, and thus it is lossless. Meanwhile, the data to be encrypted is reduced to three blocks, which are generalized bitstreams. Bitstream encryption algorithms that do not produce ciphertext expansion can be applied, and it operates independently of the other components in the paradigm, making it universal. Finally, a prototype of this paradigm is constructed, using the same encryption algorithm as in the previous scheme for comparative experiments. Two 3D object datasets are used to conduct the experiments and results demonstrate that it achieves lossless decryption and has a time advantage, while also undergoing security analysis and verification against smoothing attacks to confirm the security. © 2025 IEEE. All rights reserved,

关键词： Differential privacy

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：